{ "sessions": [ { "id": "1283926", "title": "Registration + Badge Pick-Up", "description": null, "startsAt": "2026-09-07T08:00:00", "endsAt": "2026-09-07T17:15:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [], "categoryItems": [ 485107 ], "questionAnswers": [], "roomId": 85877, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1283930", "title": "Cloakroom", "description": null, "startsAt": "2026-09-07T08:00:00", "endsAt": "2026-09-07T17:15:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [], "categoryItems": [ 485107 ], "questionAnswers": [], "roomId": 88309, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1312720", "title": "Open Computing: Building Open Software for Diverse Hardware Hosted by FlagOS", "description": "From September 7–9, 2026, KubeCon + CloudNativeCon + OpenInfra Summit + PyTorch Conference China will be co-located in Shanghai, expected to draw over 1,000 developers, architects, and technical decision-makers, alongside 400+ participating companies. As a flagship co-located event, Open AI Computing: Building an Open-Source Software Ecosystem for Diverse Heterogeneous Hardware will be held on September 7. It's a high-caliber platform for AI framework developers, compiler and runtime researchers, open-source infrastructure builders, and hardware architects to exchange ideas.\r\n\r\n\r\nAI chip diversification is now an industry reality, yet a fragmented software stack ecosystem has become the critical bottleneck preventing compute from scaling into large-model and agent applications. Every new chip demands repeated porting of operators, compilers, and inference engines; every framework must independently contend with the long tail of multiple backends. In this multi-chip era, a unified AI system software stack is essential. FlagOS is an open-source AI system software stack for multi-chip scenarios, spanning a unified operator library, a multi-backend AI compiler, training and inference frameworks, a communication library and other core components. As the host of this event, the FlagOS community will bring together the most representative experts from Beijing Academy of Artificial Intelligence (BAAI), Shanghai AI Laboratory, PyTorch Foundation, vLLM, SGLang, NVIDIA,etc. to jointly explore the technical roadmaps and collaboration models of an open AI system software stack—showcasing new system software technologies that span multiple chip architectures, and discussing hardware–software co-optimization approaches across emerging AI chip architectures, open-source compilers and acceleration frameworks.\r\n\r\nFor questions regarding the event, please contact: contact@flagos.io \r\n\r\nRegistration will open by August 26, 2026.", "startsAt": "2026-09-07T13:00:00", "endsAt": "2026-09-07T17:00:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [], "categoryItems": [ 503598 ], "questionAnswers": [ { "questionId": 134034, "answerValue": "http://bagevent.com/event/flagos" } ], "roomId": 85688, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1308451", "title": "OSPOlogy + OSPO Summit China 2026 | [Pre-Registration Required]", "description": "The OSPOlogy + OSPO Summit China 2026 focuses on the evolution of Open Source Program Management (OSPO) functions and corporate open-source governance in the era of Agentic AI. The conference will explore key topics including how Agentic AI supports practical OSPO activities, Large Language Model (LLM) and data governance, AI-driven software supply chains and enterprise global expansion strategies as well as open strategies and innovation mechanisms.\r\n\r\nDon't miss out on this CNCF-hosted co-located event, register today! https://www.lfopensource.cn/kubecon-cloudnativecon-openinfra-summit-pytorch-conference-china/co-located-events/ospology-ospo-summit/#registration-details", "startsAt": "2026-09-07T13:30:00", "endsAt": "2026-09-07T18:00:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [], "categoryItems": [ 502427 ], "questionAnswers": [ { "questionId": 134034, "answerValue": "https://www.lfopensource.cn/kubecon-cloudnativecon-openinfra-summit-pytorch-conference-china/co-located-events/ospology-ospo-summit/" } ], "roomId": 88631, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1312708", "title": "AI Native Open Session Hosted by ETSI ISG NFV", "description": "Please note that this is an off-site Sponsor-hosted Co-located event.\r\n\r\nLocation: Pudong Shangri-La, Shanghai\r\n\r\nCNCF has led the revolution of cloud native for the IT industry, and telecommunications have followed this direction since CNCF issued the definition of cloud native. Organizations like ETSI NFV ISG have aligned with CNCF’s flagship direction, and global telecom operators have deployed their 5G networks with a cloud‑native vision.\r\n\r\nWith the rapid growth of AI and agent technologies, cloud‑native platforms offer a scalable, reliable, and standardized foundation for modern applications. CNCF is proactively defining and leading the AI‑Native paradigm through open standards and projects, ensuring future software development efficiently leverages artificial intelligence at scale.\r\n\r\nETSI NFV ISG held the first AI‑Native Open Session at KubeCon + CloudNativeCon Europe 2026 and encouraged CNCF to continue leading this direction. On 7 September 2026, during KubeCon + CloudNativeCon + OpenInfra Summit + PyTorch + AgntCon China 2026, ETSI NFV ISG will organize the 2nd AI Native Open Session, gathering global AI‑native leaders in Shanghai.\r\n\r\nETSI NFV ISG held the first AI native summit and encouraged CNCF to continue to lead ‘AI Native’ direction at Kubecon EU Amsterdam 2026.\r\n\r\nAttendance is free of charge.\r\n\r\nFor questions regarding the event, please contact: denghui02@hotmail.com", "startsAt": "2026-09-07T13:30:00", "endsAt": "2026-09-07T18:30:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [], "categoryItems": [ 503598 ], "questionAnswers": [ { "questionId": 134034, "answerValue": "https://nfvwiki.etsi.org/index.php?title=2nd_AI_Native_Open_Session_Hosted_by_ETSI_ISG_NFV_%E2%80%93_7_Sep_2026_%E2%80%93_Shanghai" } ], "roomId": 88856, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1312719", "title": "Hardware-Aware AI: Building PyTorch Workloads Across Accelerators Hosted by Huawei", "description": "Driven by the vigorous growth of diverse computing resources, AI workloads are required to run seamlessly on a wide range of heterogeneous accelerators such as GPUs, NPUs and XPUs. Centered on hardware‑aware AI, this forum focuses on building the cross‑accelerator ecosystem for PyTorch. It shares hands‑on industry practices covering framework‑level adaptation, operator implementation, compiler co‑optimization, workload migration and performance tuning, and jointly explores the practical challenges and future technical pathways for integrating domestic heterogeneous computing into the upstream PyTorch community.\r\n\r\nFor questions regarding the event, please contact: zesheng.zong@outlook.com \r\n\r\nRegistration will open by August 26, 2026", "startsAt": "2026-09-07T14:00:00", "endsAt": "2026-09-07T16:00:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [], "categoryItems": [ 503598 ], "questionAnswers": [], "roomId": 80137, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1314315", "title": "PyTorch Conference Community Night, Sponsored by Huawei", "description": "Location: Puzhixing, F4, Pujiang Hall\r\n(No. 1 Dongchang Road, Pudong New Area, Shanghai)\r\n\r\nJoin fellow developers, researchers, and AI builders for a fun evening of great conversation, food, and drinks. It’s the ultimate way to gear up with the community before the main conference begins tomorrow.\r\n\r\nSpace is limited, available on a first come, first entered basis. KubeCon + CloudNativeCon + OpenInfra Summit + PyTorch Conference China badge required for entry and must be picked up from the conference venue, Shanghai International Convention Center.\r\n\r\nRSVP: https://linuxfoundation.research.net/r/PyTorchReception", "startsAt": "2026-09-07T17:30:00", "endsAt": "2026-09-07T20:30:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [], "categoryItems": [ 485111 ], "questionAnswers": [], "roomId": 88856, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1255197", "title": "Registration + Badge Pick-Up", "description": null, "startsAt": "2026-09-08T07:30:00", "endsAt": "2026-09-08T18:30:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [], "categoryItems": [ 485107 ], "questionAnswers": [], "roomId": 85877, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": false }, { "id": "1255200", "title": "Cloakroom", "description": null, "startsAt": "2026-09-08T07:30:00", "endsAt": "2026-09-08T19:30:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [], "categoryItems": [ 485107 ], "questionAnswers": [], "roomId": 88309, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": false }, { "id": "1308499", "title": "From Open Source Adoption to Open Source Innovation: China’s Next Chapter", "description": "Drawing on decades of experience advancing open source in China, Professor Lu Shouqun reflects on the evolution of China’s open source ecosystem and its growing connection to the global community. He explores how open source has progressed from technology adoption to community-driven innovation and why collaboration across borders, projects, and emerging technologies will be increasingly important as AI reshapes the industry.", "startsAt": "2026-09-08T09:00:00", "endsAt": "2026-09-08T09:02:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "0150521f-3d02-4425-a1a1-707febc575e9" ], "categoryItems": [ 482870, 447045 ], "questionAnswers": [], "roomId": 80136, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1304838", "title": "Open Source for the AI Era ", "description": "This opening will introduce the complete open source AI stack, from models and frameworks to inference, cloud native operations, infrastructure and hardware, and set the stage for the stories that follow. Through real-world examples across both days, the audience will see how these technologies work together to build, run and scale frontier intelligence—and how open collaboration across the stack is shaping what comes next.", "startsAt": "2026-09-08T09:02:00", "endsAt": "2026-09-08T09:07:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "af7b6fb1-4084-4706-a495-3fc1793d97f2", "a50dbbe3-5e92-44ca-9e40-c1b39986cbc7" ], "categoryItems": [ 482870, 447044 ], "questionAnswers": [], "roomId": 80136, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1313048", "title": "Beyond Multimodal: Building Full‑Modal AGI via Open‑Weight", "description": null, "startsAt": "2026-09-08T09:07:00", "endsAt": "2026-09-08T09:10:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "82706269-934e-470b-a08c-ac8e33709d7b" ], "categoryItems": [ 482870, 447044 ], "questionAnswers": [], "roomId": 80136, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1312064", "title": "Operating Frontier Intelligence at Scale", "description": "\"Building a model is only the beginning. When AI moves into production, the challenge shifts to making every GPU count, scaling dynamically with demand, maintaining reliability, and understanding increasingly complex systems. Cloud native technologies are becoming the operating layer that makes this possible.\r\n\r\nWe’ll also look at the growing role of observability as AI systems become more complex and why the next generation of AI infrastructure will depend on open technologies working together across the stack.\"", "startsAt": "2026-09-08T09:12:00", "endsAt": "2026-09-08T09:22:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "2a4df937-4b16-4b99-b27d-a07841c63fec", "a2589860-daa8-43eb-bfa7-19c5c4327034" ], "categoryItems": [ 482870, 447044 ], "questionAnswers": [], "roomId": 80136, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1317065", "title": "Tidal Auto-scaling for Training and Inference Based on Kubernetes + KEDA", "description": "CMB AI assistants exhibit distinct peak and off-peak traffic patterns. During off-peak hours, the platform allocates idle GPUs to LoRA incremental training. As peak traffic arrives, KEDA scales out vLLM inference services based on request load, prompting training tasks to save progress at secure checkpoints and proactively release capacity. Once traffic subsides, training automatically resumes from the saved checkpoints. This solution prioritizes online services to meet latency targets while continuously leveraging off-peak resources for model iteration, achieving coordinated optimization between inference SLOs and training efficiency.", "startsAt": "2026-09-08T09:24:00", "endsAt": "2026-09-08T09:30:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "d896475c-5e85-4b88-829a-153e2912b1ca", "5bccdb2d-dac4-4b1f-a0b5-a7355828e71d" ], "categoryItems": [ 482870, 447045 ], "questionAnswers": [], "roomId": 80136, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1312046", "title": "Building Frontier Intelligence in the Open", "description": "The pace of AI innovation requires an open ecosystem where researchers, developers, model builders and hardware companies can innovate together and where breakthroughs can move quickly from experimentation to production.\r\n\r\nThis keynote explores how PyTorch has become a foundation for that ecosystem, enabling intelligence to be built and optimized across a rapidly expanding landscape of models and compute. From open models to new accelerator architectures, we'll look at how an open development ecosystem expands who can build AI, how quickly they can innovate, and how that intelligence connects to the infrastructure that ultimately runs it. Huawei sponsored keynote to be included We’ll also welcome the PyTorch Foundation’s newest Platinum Member, marking a significant new investment in the open ecosystem and the future of AI hardware innovation.", "startsAt": "2026-09-08T09:32:00", "endsAt": "2026-09-08T09:42:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "d1e7e65a-bf95-44a0-97fe-440ba8a5c310" ], "categoryItems": [ 482870, 447044 ], "questionAnswers": [], "roomId": 80136, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1317073", "title": "Ascend & PyTorch: Pioneer New AI Open Ecosystem", "description": "As AI models grow larger and their architectures become increasingly sophisticated, the industrial ecosystem faces new opportunities and challenges. Drawing on eight years of opensource practice, Ascend deepens its collaborative partnership with the PyTorch community, evolving from hardware adaptation to joint research on technical standards.\r\n\r\nTogether, we explore nextgeneration AI hardwaresoftware codesign solutions, foster interoperability between Chinese and global AI technology stacks, and build an open and sustainable AI industrial ecosystem.", "startsAt": "2026-09-08T09:42:00", "endsAt": "2026-09-08T09:47:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "716a62ee-7c24-4d67-a609-1ba94bc8b7eb" ], "categoryItems": [ 482870, 447045, 508425 ], "questionAnswers": [], "roomId": 80136, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": false }, { "id": "1313965", "title": "Towards Device-agnostic PyTorch: Building Unified Infrastructure for a Multi-Backend Ecosystem.", "description": "AI computing has moved from chips to platforms, and now to ecosystems. Cambricon's decade-long journey has proven that scale demands full-stack co-optimization—and we are taking that expertise upstream. With an upstream-first philosophy and a track record of real contributions, we are hardening PyTorch's device-agnostic foundation for every PrivateUse1 backend. Our goal: fixes upstream, seamless everywhere. Device-agnostic is just the enabler; the real prize is unlocking what each backend does best—broader device reach and richer backend-native capabilities.", "startsAt": "2026-09-08T09:47:00", "endsAt": "2026-09-08T09:52:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "75419934-433e-4c5f-8e7e-277a2c12f39f" ], "categoryItems": [ 482870, 447044, 508425 ], "questionAnswers": [], "roomId": 80136, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1311164", "title": "Inside vLLM: Production Best Practices, Model Integration and Road Map", "description": "Take a deep dive into the vLLM ecosystem to see how to optimize, scale, and extend your LLM serving infrastructure. Beyond essential production deployment strategies, we will unpack major platform updates, including the Rust frontend, KV connector, and reinforcement learning rollouts. Finally, learn the practical mechanics of integrating brand-new architectures, with Kimi K3 as an example.", "startsAt": "2026-09-08T09:54:00", "endsAt": "2026-09-08T09:57:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "378341b9-6f6f-420a-aff7-63128c669668" ], "categoryItems": [ 482870, 447045 ], "questionAnswers": [], "roomId": 80136, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1313959", "title": "PD Disaggregation vLLM Deployment on Alternative AI Accelerators Using llm-d", "description": null, "startsAt": "2026-09-08T09:59:00", "endsAt": "2026-09-08T10:04:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "0a3b95c9-939c-4904-b9b5-c2e72573beef", "32272b85-b0be-43ab-8267-b6263d4c2b35" ], "categoryItems": [ 482870 ], "questionAnswers": [], "roomId": 80136, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1309269", "title": "From Containers to Agents: The Next Cloud Native ", "description": "Every major computing shift creates a new unit of work.\r\nVirtual machines gave us infrastructure automation. Containers gave us portable applications. Kubernetes gave us a control plane for distributed systems.\r\nAgents may become the next unit of work.\r\n\r\nAn agent is not just an LLM call. It has identity, memory, goals, tools, permissions, execution state, and the ability to affect the outside world. It can fail in ways that traditional software does not: by taking the wrong action, using the wrong tool, leaking context, exceeding its authority, or behaving differently as its models and environment evolve.\r\n\r\nThe industry is rapidly building agent demos. The harder problem is operating agents in production.\r\n\r\nThis keynote proposes a cloud native view of agent infrastructure: agents as first-class workloads with declarative configuration, policy-driven tool access, observable decision paths, isolated execution, durable state, continuous evaluation, and human-in-the-loop controls.\r\n\r\nRather than treating agent infrastructure as a proprietary layer above Kubernetes, we have an opportunity to apply the lessons of cloud native computing—open interfaces, reconciliation loops, composable primitives, and policy as code—to make autonomous systems safer and more portable.\r\n\r\nThe question is no longer whether agents will run in production. The question is whether we will build the infrastructure needed to trust them there.\r\n", "startsAt": "2026-09-08T10:06:00", "endsAt": "2026-09-08T10:11:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "8b0bcad4-0594-4ae8-a240-891d6b34e55f" ], "categoryItems": [ 482870, 447045 ], "questionAnswers": [ { "questionId": 134032, "answerValue": "https://sessionize.com/download/iicosna~HupukHSbEyx51Co8oHcD6Y.pdf~kubeconch26-keynote-from-containers-to-agents-the-next-cloud-native-abstraction.pdf" } ], "roomId": 80136, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1308539", "title": "What AI Agents Need from Open Infrastructure", "description": "AI agents create a demanding kind of workload. They generate and run code, interact with external systems, and often need an isolated execution environment for only a few minutes.\r\nThe cloud native and open infrastructure communities have spent years building the foundations this workload now needs. This keynote examines how those existing capabilities can be assembled into an agent runtime, supported by new ecosystem evidence from CNCF and OpenInfra.\r\n\r\nThe idea becomes concrete through an open-source runtime stack. Kubernetes Agent Sandbox manages the sandbox lifecycle, while Kata Containers provides workload isolation, with other cloud native components supporting execution and resource delivery. Together, these existing technologies form an agent runtime designed for the needs of agentic AI.", "startsAt": "2026-09-08T10:13:00", "endsAt": "2026-09-08T10:18:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "8fee27cf-0805-4b12-91d5-350935a3c6ac" ], "categoryItems": [ 482870, 447044 ], "questionAnswers": [], "roomId": 80136, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1309274", "title": "Building an Agent Runtime with Open Infrastructure", "description": "An agent runtime has to create a working environment on demand, isolate code the platform may not trust, and release resources when the task ends. Much of this machinery already exists across cloud native and open infrastructure projects.\r\n\r\nThis keynote follows one agent task through an open-source runtime stack. \r\nKubernetes Agent Sandbox manages its lifecycle, while Kata Containers provides a dedicated guest-kernel boundary. The rest of the chain handles container execution and efficient delivery of images and model artifacts. Seen together, these projects offer a practical foundation for secure, short-lived agent workloads and a shared path for the ecosystem to improve it.\r\n", "startsAt": "2026-09-08T10:20:00", "endsAt": "2026-09-08T10:25:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "8fee27cf-0805-4b12-91d5-350935a3c6ac", "05433fa4-5ae1-424b-a513-28d3bb92ae63" ], "categoryItems": [ 482870, 447045 ], "questionAnswers": [], "roomId": 80136, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1304846", "title": "Closing Remarks", "description": null, "startsAt": "2026-09-08T10:25:00", "endsAt": "2026-09-08T10:30:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [], "categoryItems": [ 482870, 447044 ], "questionAnswers": [], "roomId": 80136, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1255188", "title": "Coffee Break ☕", "description": null, "startsAt": "2026-09-08T10:30:00", "endsAt": "2026-09-08T11:00:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [], "categoryItems": [ 485106 ], "questionAnswers": [], "roomId": 85878, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": false }, { "id": "1319134", "title": "Gold Sponsor In-Booth Demos", "description": "Sponsor: LF Open Source Software Academy (LFOSSA)\r\nDemo: Master AI & Open Source Skills and to Advance Your Career in the Age of AI\r\nBooth Number: G5\r\n\r\n\r\n\r\nIn order to facilitate networking and business relationships at the event, you may choose to visit a third party’s booth or access sponsored content. You are never required to visit third party booths or to access sponsored content. When visiting a booth or participating in sponsored activities, the third party will receive some of your registration data. This data includes your first name, last name, title, company, address, email, standard demographics questions (i.e. job function, industry), and details about the sponsored content or resources you interacted with. If you choose to interact with a booth or access sponsored content, you are explicitly consenting to receipt and use of such data by the third-party recipients, which will be subject to their own privacy policies.", "startsAt": "2026-09-08T10:30:00", "endsAt": "2026-09-08T11:00:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [], "categoryItems": [ 500857 ], "questionAnswers": [], "roomId": 85878, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1311154", "title": "Project Pavilion | Tuesday AM Project Tables ", "description": "Tuesday AM Project Tables | 10:30 - 14:30\r\n\r\nT-1\tHAMi\r\nT-2\tKarmada\r\nT-3\tArgo\r\nT-4\tDragonfly\r\nT-5\tCNCF Table\r\nT-6\tCommunity-Led Events\r\nT-7 Cilium\r\nT-8\tKmesh\r\nT-9\tFluid\r\nT-10 Atlantis\r\nT-11 StarlingX (OpenInfra project) \r\nT-11 Zuul (OpenInfra project) \r\nT-11 OpenStack (OpenInfra project)\r\nT-12 PyTorch\r\nT-13 LWS\r\n\r\nView the full project table directory here: https://www.lfopensource.cn/kubecon-cloudnativecon-openinfra-summit-pytorch-conference-china/features-add-ons/project-engagement/#project-table-directory", "startsAt": "2026-09-08T10:30:00", "endsAt": "2026-09-08T14:30:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [], "categoryItems": [ 485109 ], "questionAnswers": [], "roomId": 85878, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1255194", "title": "Solutions Showcase", "description": null, "startsAt": "2026-09-08T10:30:00", "endsAt": "2026-09-08T19:00:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [], "categoryItems": [ 485110 ], "questionAnswers": [], "roomId": 85878, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": false }, { "id": "1311104", "title": "Women's Community Gathering", "description": "Grab coffee, tea, and a snack during the morning break and meet other women, non-binary, and allies to network, connect and socialize. Look for the Reserved Sign on a table in the Solutions Showcase. \r\n\r\nWomen’s Gatherings enable meaningful networking, peer mentorship, collaboration, and sustained engagement. ", "startsAt": "2026-09-08T10:30:00", "endsAt": "2026-09-08T11:00:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [], "categoryItems": [ 485111 ], "questionAnswers": [], "roomId": 85878, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1225136", "title": "8 Million Tasks Daily: Driving Efficiency in Multi-Tenant Data Platforms at Horizon Robotics", "description": "As autonomous driving shifts to end-to-end models, Horizon Robotics faces unprecedented scale. Thousands of concurrent, sub-minute workflows pushed infrastructure to its limits, where traditional orchestration and \"Informer Choke\" caused severe instability and resource waste.\r\nWe re-architected our platform to handle 8 million daily tasks, saving 1,000+ CPU nodes. We will dive into:\r\n1. Orchestration Re-engineering: Replacing per-workflow agents with a centralized Argo Workflows Agent Service. This \"Agentless\" approach cut infrastructure costs by 2.4M CNY/month and saved 1,000+ CPU nodes.\r\n2. Scheduling & State Optimization: Overcoming Informer bottlenecks by refactoring metadata, reducing VolcanoJob size by 30%, and slashing sync latency from 4 minutes to <10 seconds.\r\n3. Stability & Multi-Tenancy: Leveraging Kueue and Volcano to implement elastic quotas and resolve preemption conflicts, ensuring system stability in large-scale multi-tenant environments.", "startsAt": "2026-09-08T11:00:00", "endsAt": "2026-09-08T11:30:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "06f510fe-e4e0-4482-8675-88959f9a2282", "975d742e-38fd-4b5f-a244-90b146c587a3" ], "categoryItems": [ 447048, 446827, 447045 ], "questionAnswers": [ { "questionId": 134032, "answerValue": "https://sessionize.com/download/ccagmeh~Sho5sytNjC4agwxxVJ33VD.pdf~kubecon_shanghai_2026_horizon_robotics.pdf" } ], "roomId": 80134, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1210254", "title": "Accelerating Sequential Recommender Models in the PyTorch Ecosystem", "description": "Sequential recommendation models share a natural connection with large language models in use of self-attention, yet production recommendation systems introduce meaningful differences that are often underappreciated.\r\n\r\nWe begin with SASRec, one of the earliest Transformer decoder-based sequential recommenders, walking through its rewrite from TensorFlow to PyTorch - highlighting how PyTorch reduced modeling complexity and surfacing the gaps between academic prototypes and production systems.\r\n\r\nWe then examine HSTU, an attention operator for sequential recommendation at scale: what makes its attention pattern structurally different from standard Transformer attention, and key GPU insights - kernel fusion, variable-length sequences, and PyTorch integration via FBGEMM.\r\n\r\nWe close with a reflection on sequential recommendation and LLMs from modeling and systems perspectives, sharing observations on emerging trends and optimization strategies across recommendation workloads.", "startsAt": "2026-09-08T11:00:00", "endsAt": "2026-09-08T11:30:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "d455f371-8318-45d0-a3be-c10b0af4bee1", "c28d10bb-0c49-4886-acca-d8dd5de44b37" ], "categoryItems": [ 447055, 446827, 447045 ], "questionAnswers": [ { "questionId": 134032, "answerValue": "https://sessionize.com/download/izuxis~nWSBcPQMv3MwVn8Qb4baGS.pdf~pytorchchina2026seqrecopt.pdf" }, { "questionId": 134033, "answerValue": "https://github.com/pmixer/SASRec.pytorch/" }, { "questionId": 134034, "answerValue": "https://github.com/pytorch/FBGEMM/commit/e0dec03942b7373d9d714c70bfc77e3b44a694e8" }, { "questionId": 134035, "answerValue": "https://github.com/NVIDIA/cudnn-frontend/tree/develop/python/cudnn/hstu_attention" }, { "questionId": 134036, "answerValue": "https://cseweb.ucsd.edu/~jmcauley/pdfs/icdm18.pdf" } ], "roomId": 80135, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1224992", "title": "Declarative Underlays: Scaling Purpose-Built Infrastructure(Clusters) for OpenStack with Cluster API", "description": "OpenStack and k8s were once considered separate ecosystems, but today many operators run OpenStack on top of Kubernetes to get the benefits of both worlds.\r\n\r\nWe have taken that integration a step further by leveraging Cluster API (CAPI) \r\nto declaratively create/manage clusters, each tuned for a specific need (hardware/software) \r\n(AI clusters /w GPUs, high-speed networking cluster, dedicated storage, etc.). \r\n\r\nThe underlay becomes a set of clusters that can be re-balanced on demand, \r\ndelivering true cloud-native agility to the physical layer.\r\n\r\nIn this session, we will focus on production-proven architecture that:\r\n- Declarative underlay clusters – repeatable, version-controlled provisioning.\r\n- Isolates specialized hardware – Cluster for (AI, Compute, Network) without interfering with other workloads\r\n- Dynamically rebalances resources – migrate idle nodes (GPUs) between clusters\r\n- Integrates OpenStack services\r\n- Operates at scale (Multi Region) – lessons learned and best-practice patterns ", "startsAt": "2026-09-08T11:00:00", "endsAt": "2026-09-08T11:30:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "6545c39f-4dd2-4489-b230-c723558d0610", "73b45691-d9d8-4744-8755-f9b7fac55079" ], "categoryItems": [ 447053, 446824, 447044 ], "questionAnswers": [ { "questionId": 134032, "answerValue": "https://sessionize.com/download/iktariq~RSKSn2UTCwcWzwsfmdNFr2.pdf~declarative_underlays_kubecon_china_2026.pdf" } ], "roomId": 80136, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1224095", "title": "📚 Building Portable Large Models on Heterogeneous AI Accelerators with FlagOS", "description": "Porting large models across diverse accelerators and running them efficiently remains difficult. Developers can download LLMs and multimodal models such as Qwen, MiniMax, and DeepSeek, yet the runtime path below PyTorch, vLLM, or SGLang is still fragmented across vendor libraries, compiler stacks, communication backends, and framework patches.\r\nThis tutorial introduces how to port large models onto various AI accelerators by using the FlagOS open-source stack. we will walk through the path from PyTorch operators to training and inference with vLLM and SGLang.\r\nAudiences will learn how Triton enables portable kernel programming, how FlagGems provide PyTorch operator coverage, how unified plugins of FlagOS for vLLM, Megatron, VeRL etc. support training and inference on different AI accelerators, and KernelGen help auto-coding for different accelerators. This tutorial shows how developers can bring open models to more hardware, with less porting effort and stronger community collaboration.", "startsAt": "2026-09-08T11:00:00", "endsAt": "2026-09-08T12:15:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "b5c46ce4-07b8-450f-a296-f7f37df30671", "16f68e1f-cee0-4a9c-ae4f-e6dcb34a31db", "04ec9d62-8714-419d-90a2-2fd6cea2045a" ], "categoryItems": [ 485113, 446827, 447044, 485134 ], "questionAnswers": [], "roomId": 80137, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1260073", "title": "Project Lightning Talk: Opening Remarks", "description": null, "startsAt": "2026-09-08T11:00:00", "endsAt": "2026-09-08T11:05:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "12dfc79a-ac6a-463e-8229-6429ce9a29e5" ], "categoryItems": [ 485109 ], "questionAnswers": [], "roomId": 85688, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": false }, { "id": "1223701", "title": "Project Lightning Talk: Architecture Without Gatekeepers: Decoupling for Neutrality", "description": "Open source only works if the community can contribute without hitting a wall. We realized our architecture at OpenEverest was unintentionally biased toward specific tools, so we rebuilt it. I’ll share how we moved to a modular plugin system that decouples our core from specific operators. By supporting everything from Valkey to competing Postgres flavors, we’ve eliminated vendor lock-in and ensured the community—not the architecture—dictates the project's future.", "startsAt": "2026-09-08T11:07:00", "endsAt": "2026-09-08T11:12:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "b14e2c8e-b7da-4835-9a2c-7c9705ac94dd" ], "categoryItems": [ 485109, 447044, 486113 ], "questionAnswers": [], "roomId": 85688, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1215494", "title": "Project Lightning Talk: From Static Slices to Elastic GPUs: Dynamic MIG with HAMI", "description": "In Kubernetes, NVIDIA MIG is typically pre-partitioned, forcing operators to decide GPU slice layouts before workloads arrive. This static model struggles with mixed, dynamic workloads, often leading to fragmentation and underutilization.\r\n\r\nThis talk presents a scheduling-driven approach: instead of fitting workloads into fixed slices, HAMI adapts GPU partitioning based on real-time scheduling decisions by integrating the scheduler with the device plugin.\r\n\r\nThe result is a more flexible and efficient use of GPU resources—while still operating within hardware constraints.\r\n\r\nThe key idea is simple: GPU partitioning should follow scheduling, not precede it.", "startsAt": "2026-09-08T11:14:00", "endsAt": "2026-09-08T11:19:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "0a3b95c9-939c-4904-b9b5-c2e72573beef" ], "categoryItems": [ 485109, 447045, 486119 ], "questionAnswers": [], "roomId": 85688, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1224863", "title": "Project Lightning Talk: KubeEdge Everywhere: Latest Project Update with industrial cases", "description": "Since becoming the first cloud-native edge project to graduate from the CNCF, KubeEdge adoption has exploded. From intelligent transportation and smart cities to energy and banking, it is now the backbone of critical edge-cloud ecosystems. In this lightning talk, we will highlight the latest post-graduation features, governance updates, and real-world success stories that demonstrate the power of KubeEdge for Edge AI and industrial workloads.", "startsAt": "2026-09-08T11:21:00", "endsAt": "2026-09-08T11:26:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "67f65a78-9fba-498c-84d1-ddf27bac6de3" ], "categoryItems": [ 485109, 447044, 486111 ], "questionAnswers": [], "roomId": 85688, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": false }, { "id": "1222021", "title": "Project Lightning Talk: Multi-cluster Orchestration System: Karmada Updates and Use Cases", "description": "Karmada (Kubernetes Armada) is a Kubernetes management system that enables you to run your cloud-native applications across multiple Kubernetes clusters and clouds. \r\n\r\nIn this presentation, the maintainer of the Karmada project will share: \r\n\r\n- A Brief introduction to Karmada.\r\n- New features over the last year\r\n- Real-world case studies\r\n- Overview of the community \r\n- Roadmap\r\n", "startsAt": "2026-09-08T11:28:00", "endsAt": "2026-09-08T11:33:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "1afdc8f0-ebbb-4f55-be52-64bf0815b421", "e578e9c1-d7b3-4071-bc8e-a625abaa2b45" ], "categoryItems": [ 485109, 447045, 486116 ], "questionAnswers": [], "roomId": 85688, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1218468", "title": "Project Lightning Talk: Atlantis: Terraform Pull Request Automation for Cloud Native Teams", "description": "Atlantis is a CNCF Sandbox project that provides Terraform and OpenTofu pull request automation for cloud native teams. In this 5-minute Project Lightning Talk, I will give a concise project update and show how Atlantis fits into modern infrastructure delivery: making plans visible in pull requests, keeping applies approval-gated, and preserving an auditable Git-based workflow.\r\n\r\nThe talk will focus on the core Atlantis workflow: webhook-triggered plans, pull request comments, policy-aware approvals, controlled applies, and self-hosted deployment patterns for Kubernetes-based teams. I will also briefly cover why this workflow remains important as the open source IaC ecosystem evolves around Terraform and OpenTofu.\r\n\r\nAttendees will leave with a practical understanding of what Atlantis does, when it is a good fit, recent community focus areas, and how to get involved as users or contributors.", "startsAt": "2026-09-08T11:35:00", "endsAt": "2026-09-08T11:40:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "ee1a3539-1e0f-442e-a18c-910cb38e1818" ], "categoryItems": [ 485109, 447045, 486118 ], "questionAnswers": [], "roomId": 85688, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1319491", "title": "Project Lightning Talk: Run Sandbox Securely and Cost-Effectively with OpenKruise Agents", "description": "Running sandboxes for LLM-generated code in production forces an uncomfortable three-way trade-off between security, latency, and cost. The sandbox executes untrusted model code, so it cannot hold a single outbound credential; a warm pool delivers sub-second creation but burns money at rest; and the agents inside must answer user requests and scheduled jobs instantly, even though they sit idle most of the time. This talk shows how OpenKruise Agents squares that triangle by combining horizontal and vertical elasticity on the warm pool, on-the-fly token injection through an agent gateway, and smart sandbox pause-and-resume. The takeaway: with Envoy and OpenKruise in-place update, production-grade agent sandboxes are already within reach.", "startsAt": "2026-09-08T11:42:00", "endsAt": "2026-09-08T11:47:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "a6e8b69b-7ea1-4686-a849-52d913553d69" ], "categoryItems": [ 485109, 486109 ], "questionAnswers": [], "roomId": 85688, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": false }, { "id": "1225364", "title": "Scaling Digital Employees at China Merchants Bank in Production: A Controlled AI Agents Approach", "description": "China Merchants Bank has transitioned from AI prototypes to large-scale internal deployment of Large Language Models, AI Agents for Digital Employees. To manage the inherent unpredictability of these autonomous systems in the regulated banking sector, the bank utilizes an “In-Control” architectural pattern built on Platform Engineering principles for its internal operations.\u2028This framework rests on three pillars:\u2028Platform Engineering for AI: Streamlines internal deployment using standardized “AI Agent-as-a-Service” templates to empower Digital Employees.\u2028Secure Sandboxing: Employs Kubernetes-native micro-VMs to execute agent-generated code in isolation, protecting core infrastructure.\u2028Financial Guardrails: Implements real-time monitoring and deterministic policies to ensure compliance and budget control within the enterprise.\u2028This structured approach allows the bank to manage internal Digital Employees in cloud environments while maintaining strict data sovereignty and auditability.", "startsAt": "2026-09-08T11:45:00", "endsAt": "2026-09-08T12:15:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "9936935c-dfa6-49c1-91bd-13281ab40901" ], "categoryItems": [ 447048, 446824, 447044 ], "questionAnswers": [], "roomId": 80134, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1189573", "title": "Dragonfly's CNCF Graduation Journey and Large-Scale AI Model Distribution in Practice", "description": "In 2025, after nearly eight years of incubation within CNCF, the Dragonfly project officially graduated. But getting there wasn't easy. Growing from an internal project at Alibaba to a full-community-driven CNCF graduated project came with its fair share of challenges. In this talk, we will discuss the inner workings of Dragonfly's journey to its current state.\r\n\r\nWe’ll cover our lessons in community governance, including how we evolved from a small group of maintainers to transparent decision-making and attracted global contributors through effective open-source practices. \r\n\r\nOn the technical side, the increasing popularity of LLMs has made the efficient distribution of model files at the GB or even TB scale a significant challenge. We'll dive into how Dragonfly's P2P capabilities have evolved to tackle the challenge of pulling massive model files concurrently across Kubernetes clusters, along with best practices we've developed for AI inference and training workloads.", "startsAt": "2026-09-08T11:45:00", "endsAt": "2026-09-08T12:15:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "1413a1ce-360d-4c56-8dd5-c820b1cd9aa9", "13b40d9b-27d7-42f1-a074-21e7e84a442f" ], "categoryItems": [ 447055, 446827, 447045 ], "questionAnswers": [], "roomId": 80135, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1185605", "title": "Exploring HiFloat8: A Tapered Format Complementing the FP8 Ecosystem for Robust Model Training", "description": "Standard FP8 formats suffer from frequent gradient overflows and heavy reliance on complex Delayed Scaling, which often lead to training instability or suboptimal convergence in large models. This session introduces HiFloat8 (HiF8) — a tapered precision format that offers an alternative approach to managing dynamic range. This \"natural\" alignment with neural network weight/gradient distributions allows HiF8 to capture high-magnitude outliers without the aggressive scaling required by standard FP8.We explore how HiF8 can works in the training and inference procedure.We will demonstrate the implementation of HiF8 within the ecosystem on vllm and\r\ndeepspeed. They allow developers to evaluate performance of Hif8 on GPUs. Also, we will give an analysis of training stability and final loss parity where HiF8 provides relatively the same accuracy and 1.5-1.7 times GEMM performance than FP16. Finally, we will share insights from our ongoing collaboration on dedicated hardware support for HiF8.", "startsAt": "2026-09-08T11:45:00", "endsAt": "2026-09-08T12:15:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "3f378b4a-b718-4c9f-ab48-07c8fc1df1d1" ], "categoryItems": [ 447046, 446826, 447045 ], "questionAnswers": [ { "questionId": 134032, "answerValue": "https://sessionize.com/download/irpairze~T3bdM6bbezC1rtgv8nPTWZ.pdf~pytorchcon-china.pdf" } ], "roomId": 80136, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1319502", "title": "Project Lightning Talk: Let's Talk About Perses, the Open Specification for Prometheus and More", "description": "In this talk we will discuss perses a CNCF sandbox project, which is an open specification for Dashboards, perses can help you organize your data-sources, has embeddable components is devops ready and Kubernetes native. I'll be sharing what the Project is and what the latest is on this project in this short talk along with a small demo. ", "startsAt": "2026-09-08T11:49:00", "endsAt": "2026-09-08T11:54:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "f7834264-1b08-4f5a-874f-e0654a5a3016" ], "categoryItems": [ 485109, 506816 ], "questionAnswers": [], "roomId": 85688, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1322749", "title": "Project Lightning Talk: Let’s Talk About Headlamp, the Extensible Open-Source Kubernetes UI", "description": "In this talk, we will explore Headlamp, a CNCF Sandbox project that provides a modern and user-friendly interface for Kubernetes. The session will cover what Headlamp is, its key features and extensibility, recent developments, and how it fits into the Kubernetes ecosystem. This short talk will provide an introduction to the project and what to know when getting started.", "startsAt": "2026-09-08T11:53:00", "endsAt": "2026-09-08T11:58:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "b7541db4-ccd8-42f8-8d81-53a75ae4e07c" ], "categoryItems": [ 485109, 507394 ], "questionAnswers": [], "roomId": 85688, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": false }, { "id": "1260075", "title": "Project Lightning Talk: Closing Remarks", "description": null, "startsAt": "2026-09-08T11:58:00", "endsAt": "2026-09-08T12:00:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "12dfc79a-ac6a-463e-8229-6429ce9a29e5" ], "categoryItems": [ 485109 ], "questionAnswers": [], "roomId": 85688, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": false }, { "id": "1255193", "title": "Lunch (Not Provided) 🍜", "description": null, "startsAt": "2026-09-08T12:15:00", "endsAt": "2026-09-08T13:45:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [], "categoryItems": [ 485106 ], "questionAnswers": [], "roomId": 88856, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": false }, { "id": "1177779", "title": "Architecting Secure Agentic Workflows on Kubernetes: A Financial Sector Case Study", "description": "As organizations move beyond basic LLM integrations toward autonomous agentic workflows, the infrastructure required to support these systems grows increasingly complex. Running multi-agent architectures in production introduces unique challenges around tool discovery, secure access, and traffic routing.\r\n\r\nThis session explores how to leverage Kubernetes and cloud-native abstractions to develop, test, and deploy AI agents at scale. Using a reference architecture leveraging the Kagenti project, we will demonstrate how to construct a secure, scalable agentic environment. The talk covers unifying tool access via an MCP Gateway, enforcing zero-trust workload identity with SPIFFE/SPIRE, and standardizing inter-agent communication.\r\n\r\nTo illustrate these concepts, we will walk through a real-world financial use case: an autonomous agent that securely accesses market data and executes simulated transactions using financial tools over MCP, demonstrating end-to-end cloud-native deployment.", "startsAt": "2026-09-08T13:45:00", "endsAt": "2026-09-08T14:15:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "0ed2ceb9-abe1-4de0-ba1d-64da3a79ad5f", "72e83064-d18e-4278-9bc6-2a3e43549995" ], "categoryItems": [ 447046, 446826, 447044 ], "questionAnswers": [], "roomId": 80134, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1221856", "title": "Zero-Trust Traffic Governance For Kubernetes AI Agent Sandboxes", "description": "When AI agents run in Kubernetes, network access can no longer be treated as a static allowlist problem. An agent may execute generated code, choose tools at runtime, call external model or API endpoints, and carry credentials that should not be exposed to the sandbox itself. In our sandbox platform, we treated this as a zero-trust traffic governance problem: every outbound request needs a clear destination, identity, credential boundary, and audit trail.\r\n\r\nThis talk shares how we built that control path for AI agent sandboxes. We will cover how sandbox traffic is restricted by domain, service, and network target; how platform-wide guardrails and tenant-level allowlists are combined; and how DNS changes, first-packet timing, and short-lived sandboxes affect the design. We will also show how the same path handles identity injection, token replacement, LLM request audit, and forced forwarding through an internal model gateway, without putting a heavy sidecar next to every sandbox.", "startsAt": "2026-09-08T13:45:00", "endsAt": "2026-09-08T14:15:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "f5200c5c-d81a-488c-b57a-520e94c626be", "ea195666-322a-48a9-9402-c6a944a4c4b1" ], "categoryItems": [ 447052, 446826, 447045 ], "questionAnswers": [ { "questionId": 134032, "answerValue": "https://sessionize.com/download/ioyjayxe~SGt9nR1NtS54hTts9Bb8dy.pdf~zero-trust_traffic_governance_for_kubernetes_ai_agent_sandboxes.pdf" } ], "roomId": 80135, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1224503", "title": "From Chatbots to Agentic AI: Running NVIDIA Dynamo the Kubernetes Way", "description": "Agentic AI is redefining inference on Kubernetes. Long tool-call workflows create sustained KV-cache pressure; parallel subagents need shared prefixes and consistent routing; and sessions must survive pod/GPU failures without losing context. Prefill–decode disaggregation is becoming a baseline, while multimodal pipelines push architectures toward multi-stage EPD-style decomposition and more stateful coordination.\r\n\r\nThis talk shares production lessons from running agentic inference on Kubernetes with NVIDIA Dynamo for cache-aware routing and disaggregated serving, and RoleBasedGroup (RBG) for stateful continuity, discovery, and role/unit coordination across pipeline stages. We compare deployment patterns (Dynamo-native vs RBG), what broke under cache contention and partial failures, and how we modeled routing, session continuity, recovery, and observability in Kubernetes-native ways.", "startsAt": "2026-09-08T13:45:00", "endsAt": "2026-09-08T14:15:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "3fdf03ac-784d-4055-870d-f48967a60b89", "150879da-174d-4ebf-aa1c-ebebdbd5e20d" ], "categoryItems": [ 447046, 446827, 447044 ], "questionAnswers": [], "roomId": 80136, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1224567", "title": "📚 Advanced NCCL Tutorial: Beyond Collectives", "description": "In this tutorial, we will teach you how to use the newest features in NCCL: symmetric memory and GPU-initiated networking (GIN), which are used in state-of-the-art mixture-of-experts (MoE) implementations like DeepEP v2.\r\n\r\nThis tutorial assumes you know how to initialize NCCL and perform common host-initiated operations. We will explain the concept of symmetric memory and how to use it to scale communication within an NVLink domain. We will also show you how to scale across interconnected GPU nodes using GIN. The examples in the hands-on portion will be simple, but sufficient to demonstrate the concepts and allow attendees to begin working towards complex use cases.\r\n\r\nAll software materials for this course are open source. \r\n\r\nContent for this tutorial will be available in both English and Chinese. The instructors include both native English and Chinese speakers. We will prepare to teach the course in both languages and select based upon the preference of the attendees.\r\n\r\n", "startsAt": "2026-09-08T13:45:00", "endsAt": "2026-09-08T15:00:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "4f75f000-e518-49a6-b3a1-0d79346c5220", "a1d5009a-4f79-4103-a478-67eba7b40c43", "f9b10346-995d-4c20-a3a4-fdabef35fb13" ], "categoryItems": [ 485113, 446824, 447044, 485134 ], "questionAnswers": [ { "questionId": 134032, "answerValue": "https://sessionize.com/download/okugpa~ErvMRkjRjLFiWksGeRKUuR.pdf~pytorch-china-2026-tutorial.pdf" }, { "questionId": 134033, "answerValue": "https://github.com/jeffhammond/pytorch-china-2026-nccl-tutorial" }, { "questionId": 134036, "answerValue": "https://docs.nvidia.com/deeplearning/nccl/user-guide/docs/" } ], "roomId": 80137, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1222146", "title": "Beyond Model Sharding: Atomic Scheduling and Disaggregated LLM Serving with LeaderWorkerSet", "description": "LLM serving on Kubernetes is moving beyond model sharding. In production, one inference replica may be a coordinated group of pods across nodes, GPUs, and network topology. If only part of the group is scheduled, accelerators can be reserved without serving traffic. With disaggregated prefill and decode, rollout, scaling, and placement mismatches can also hurt latency and GPU utilization.\r\nLeaderWorkerSet provides a Kubernetes-native API for managing a group of pods as one workload unit for multi-host inference. This talk focuses on two recent areas of LWS evolution: gang scheduling and DisaggregatedSet. We will show how PodGroup integration with Volcano, scheduler-plugins, and YuniKorn schedules leader and worker pods together, and how DisaggregatedSet coordinates multiple LWS resources for prefill and decode.\r\nAttendees will learn when Deployment is no longer enough, when LWS is the right abstraction, and when multiple LWS resources should be managed as one logical service.\r\n", "startsAt": "2026-09-08T13:45:00", "endsAt": "2026-09-08T14:15:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "29eabf0a-4372-4f76-b8b9-4deb70ff2a79", "4e85174f-e168-489c-8626-d2c762b3fc99" ], "categoryItems": [ 485108, 447045, 486115 ], "questionAnswers": [ { "questionId": 134032, "answerValue": "https://sessionize.com/download/vgayseh~7WqizQY5Mfk3vudj2dh8Bh.pdf~beyond-model-sharding-atomic-scheduling-and-disaggregated-llm-serving-with-leaderworkerset.pdf" } ], "roomId": 85688, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1311081", "title": "Peer Group Mentoring", "description": "Peer Group Mentoring allows participants to meet with experienced open source veterans across many projects. Mentees are paired with 2 – 8 other people in a pod-like setting to explore technical, community, and career questions together.\r\n\r\nIf you're interested in attending as a Mentee, seats are first come first served.\r\nIf you're interested in being a Mentor, please sign up here: https://www.surveymonkey.com/r/ChinaMentor26", "startsAt": "2026-09-08T13:45:00", "endsAt": "2026-09-08T15:00:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [], "categoryItems": [ 485111 ], "questionAnswers": [], "roomId": 87761, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1224932", "title": "Spec-Driven Evolution: How AI Agents are Rewriting the OpenStack/K8s Playbook", "description": "How do you turn complex Telco standards into production-ready code—without months of manual design and validation?\r\n\r\nThis session introduces a spec-driven development workflow using AI coding agents to accelerate the design, implementation, and testing of features in OpenStack Tacker. \r\nWe show how AI agents navigate large codebases to propose architectural solutions and generate test suites directly from specifications.\r\n\r\nTwo use cases are presented: (1) Cloud-native high availability, with AI-assisted design of Kubernetes Operators enabling redundancy and fast failover toward 99.999% SLA; and (2) Massive-scale vRAN, with a Leader–Conductor style coordination model inspired by OpenStack orchestration systems coordinating thousands of VNFs/CNFs, along with automated edge-case test generation.\r\n\r\nAttendees will learn a repeatable workflow for combining AI with upstream collaboration to accelerate the delivery of carrier-grade systems while maintaining OpenInfra and Kubernetes standards.", "startsAt": "2026-09-08T14:30:00", "endsAt": "2026-09-08T15:00:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "516cce70-ba6b-406e-845c-4e0ea6471583", "0992ddcd-389f-4343-9151-68dc8c1c9165" ], "categoryItems": [ 447052, 446826, 447044 ], "questionAnswers": [], "roomId": 80134, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1212758", "title": "Behind the Scenes of Building a Gateway API Dataplane With Envoy Proxy", "description": "Kubernetes Gateway API has become the standard for traffic management in cloud-native environments, with Envoy Proxy emerging as the dominant data plane across implementations. Despite sharing the same proxy, implementations differ in how they translate Gateway API resources into Envoy configuration and distribute it via xDS.\r\n\r\nThis talk analyzes those differences in depth, covering the trade-offs between State-of-the-World and Delta xDS, how ADS ordering guarantees help achieve hitless configuration rollouts and where they still fall short, and how MuxCache can decouple reconciliation in the controller. We’ll also look at how control planes model backends and routes and what that means for scale, policy attachment, observability, and operational safety.\r\n\r\nDrawing on hands-on experience building a Gateway API data plane, you'll leave with lessons learned and actionable insights, whether you're building or operating a Gateway API implementation or evaluating one for production.\r\n", "startsAt": "2026-09-08T14:30:00", "endsAt": "2026-09-08T15:00:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "0d5dfadd-1092-4508-8d50-36e18bf685de", "1892791a-a1af-4fe9-b7c8-ee89de2c76c3" ], "categoryItems": [ 447052, 446824, 447044 ], "questionAnswers": [ { "questionId": 134032, "answerValue": "https://sessionize.com/download/plauche~WgfkVQWTwGRhuX3K9a2J4F.pdf~kubecon-china-2026-behind-the-scenes-of-building-a-gateway-api-dataplane-with-envoy-proxy.pdf" } ], "roomId": 80135, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1221808", "title": "How Intsig Serves Billions of Document Scans: GPU Virtualization at Scale with HAMi", "description": "Intsig built CamScanner (300M downloads) and runs one of China's busiest document processing platforms. Their GPU problem is unusual: they don't juggle dozens of models — they run one thing (OCR) at absurd concurrency across ~1,000 cards. The bottleneck isn't placement, it's queue time. People waiting in line.\r\n\r\nThey moved from Tencent QGPU to HAMi, the CNCF Sandbox project, and got virtualization, scheduling, and monitoring in one package. Queue times dropped. Utilization went up.\r\n\r\nThis isn't just an Intsig talk. HAMi runs in production at SF Express (GPU fleet cut from 1,400 to 1,000 cards), China Merchants Bank (10,000+ GPUs, utilization 20%→80%), NIO (10x CI efficiency), and ICBC (utilization 20%→70%). We'll share patterns that work and anti-patterns that waste money — like why slicing GPUs below 1/6 backfires.\r\n\r\nA quick Chaterm demo closes the talk — Intsig's open-source AI terminal for GPU cluster ops in plain language.", "startsAt": "2026-09-08T14:30:00", "endsAt": "2026-09-08T15:00:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "22d235b4-af0f-4f0c-a854-cdccffad169e", "5ecf3700-16a6-4832-b340-e7b3b86beca9" ], "categoryItems": [ 447046, 446826, 447044 ], "questionAnswers": [], "roomId": 80136, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1174542", "title": "Dragonfly V2.5.0 - Intro, Updates, Data Distribution in AI Infrastructure", "description": "Dragonfly provides efficient, stable, and secure file distribution and image acceleration using P2P technology within cloud-native architectures. This talk will briefly introduce Dragonfly and highlight the features of its latest version. Key updates include enhanced security and new functionalities tailored for more efficient and robust model distribution. We will also demonstrate how Dragonfly preheats and distributes AI models (packaged as OCI Artifacts) to read-only volumes in Kubernetes, enabling faster deployments. Additionally, we will introduce P2P-based state snapshot and restore capabilities in AI agent scenarios.", "startsAt": "2026-09-08T14:30:00", "endsAt": "2026-09-08T15:00:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "13b40d9b-27d7-42f1-a074-21e7e84a442f", "f4827ef5-9d40-44e9-8f70-be0c21c27af2" ], "categoryItems": [ 485108, 447045, 486106 ], "questionAnswers": [], "roomId": 85688, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": false }, { "id": "1255189", "title": "Coffee Break ☕ ", "description": null, "startsAt": "2026-09-08T15:00:00", "endsAt": "2026-09-08T15:30:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [], "categoryItems": [ 485106 ], "questionAnswers": [], "roomId": 85878, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": false }, { "id": "1311156", "title": "Project Pavilion | Tuesday PM Project Tables ", "description": "Tuesday PM Project Tables | 15:00 - 19:00 \r\n\r\nT-1\tHAMi\r\nT-2\tVolcano\r\nT-3\tArgo\r\nT-4\tDragonfly\r\nT-5\tCNCF Table\r\nT-6\tCommunity-Led Events\r\nT-7 Cilium\r\nT-8\tWasmEdge\r\nT-9\tOpenKruise\r\nT-10 K3S\r\nT-11 OpenStack (OpenInfra project)\r\nT-11 Kata Containers (OpenInfra project)\r\nT-12 vLLM (PyTorch project)\r\nT-13 LWS\r\n\r\n\r\nView the full project table directory here: https://www.lfopensource.cn/kubecon-cloudnativecon-openinfra-summit-pytorch-conference-china/features-add-ons/project-engagement/#project-table-directory", "startsAt": "2026-09-08T15:00:00", "endsAt": "2026-09-08T19:00:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [], "categoryItems": [ 485109 ], "questionAnswers": [], "roomId": 85878, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1328060", "title": "Women's Community Gathering (Additional Session)", "description": "Grab coffee, tea, and a snack during the morning break and meet other women, non-binary, and allies to network, connect and socialize. Look for the Reserved Sign on a table in the Solutions Showcase. \r\n\r\nWomen’s Gatherings enable meaningful networking, peer mentorship, collaboration, and sustained engagement. ", "startsAt": "2026-09-08T15:00:00", "endsAt": "2026-09-08T15:30:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [], "categoryItems": [ 485111 ], "questionAnswers": [], "roomId": 85878, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1219464", "title": "Minutes, Not Weeks: Long-Term Memory for AI Applications with PowerMem", "description": "Every AI app eventually needs memory. Most teams hit the same walls: choosing a backend locks them into a later rewrite, wiring retrieval and lifecycle bleeds into every layer, and non-Python teammates have no clean way in.\r\n\r\nPowerMem is an open-source (Apache 2.0) memory library that treats these as developer-experience problems. The talk covers the design choices: auto-loading config from .env so the first example is under ten lines of Python; one memory API across five surfaces (Python SDK, pmem CLI with shell, HTTP API with OpenAPI docs, MCP server over SSE/stdio/streamable-http, web dashboard); pluggable storage (OceanBase, PostgreSQL, SQLite) and model providers so prototype and production share one code path; schema migration and backup/restore baked into the CLI.\r\n\r\nAttendees leave with patterns for multi-surface developer experience and a practical checklist for evaluating memory libraries.", "startsAt": "2026-09-08T15:30:00", "endsAt": "2026-09-08T16:00:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "a5f0abdd-1eda-4470-97d7-75c7489ca5ac" ], "categoryItems": [ 447049, 446827, 447045 ], "questionAnswers": [ { "questionId": 134032, "answerValue": "https://sessionize.com/download/ilkaywef~wJQphDzZjZhp21FbPgkM5q.pptx~powermem_kubecon_2026_6.0_en.pptx" } ], "roomId": 80134, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1224322", "title": "Building AI4S Systems for Scientific Research on Kubernetes", "description": "Researchers are exploring how AI workflows can accelerate science in recent AI waves. Most existing workflows are built for AI/ML model training, and they fall short for domain-specific scientific work that demands traceable, and reproducible workflows.\r\nTo bridge this gap, China Mobile Research Institute built an AI4S system that unifies computing infrastructure and a zero-code software platform. The infrastructure uses Kubernetes to federate multiple cross-region clusters into a pool of heterogeneous resources. The platform covers the entire research lifecycle, form model training to scientific agent development, allowing researchers to build customized assistants without coding.\r\nIn collaboration with an RNA-focused biology lab, the system was applied to train a RNA structure prediction model. The resulting model achieved very well performance.\r\nThis presentation will share the system architecture, design decisions, and key lessons learned from deploying AI in real scientific research.", "startsAt": "2026-09-08T15:30:00", "endsAt": "2026-09-08T16:00:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "b39c90e5-f927-43b6-aba9-3f38853fd0cf", "bf6c7299-d682-454a-b453-80e837035988" ], "categoryItems": [ 447054, 446827, 447045 ], "questionAnswers": [ { "questionId": 134032, "answerValue": "https://sessionize.com/download/iuytanse~B1n5STzLVTwU3cPDH5uarN.pdf~building-ai4s-systems-for-scientific-research-on-kubernetes0902_v2.0.pdf" } ], "roomId": 80135, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1209151", "title": "KernelAgent: Hardware-Guided GPU Kernel Optimization via Multi-Agent Orchestration", "description": "Recently, the PyTorch team released KernelAgent, an open agentic system achieving 100% correctness across all 250 L1/L2/L3 KernelBench tasks. We extend that work by adding a hardware-guided optimization layer to the existing framework. Building on the previous correctness-focused pipeline, KernelAgent integrates GPU hardware-performance signals into a closed-loop multi-agent workflow to guide the optimization for Triton Kernels.\r\n\r\nWe evaluate the kernels generated by KernelAgent on all 100 L1 KernelBench tasks. Overall, it achieved 2.02x speedup over generated kernels from earlier versions. On average, KernelAgent generated 1.56x speedup when compared to default torch.compile, outperforming 65 of 100 KernelBench L1 tasks and achieving 89% of the hardware roofline efficiency on the H100.\r\n\r\nThe optimization codebase is located at KernelAgent repo with documentation to get started. We also share a selection of end-to-end KernelAgent optimization artifacts in the open-source repo. ", "startsAt": "2026-09-08T15:30:00", "endsAt": "2026-09-08T16:00:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "b33fc27e-2735-4035-a83f-39b7790e9e44", "bc0746bf-bfbe-44c5-bf98-42dcaaa48ac7", "2d6077f4-3da7-43b6-bd0d-eea01cef0486" ], "categoryItems": [ 447046, 446824, 447044 ], "questionAnswers": [ { "questionId": 134032, "answerValue": "https://sessionize.com/download/shabbek~3ejw5qoAvJxxfrN2Tho1NE.pptx~ptc-china-kernelagent-1.pptx" } ], "roomId": 80136, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1227863", "title": "10 Years of Cilium: Connecting, Securing, and Simplifying the Cloud Native Stack", "description": "Join us as we celebrate a decade of Cilium, now the de-facto standard CNI for Kubernetes and a cornerstone of cloud native networking and security. This session provides updates on the latest Cilium release and highlights how its unified eBPF-powered stack is transforming Kubernetes environments and beyond by replacing fragmented toolchains with seamless, secure, scalable, and simplified solutions.\r\n\r\nWe’ll showcase how Anta Group used the 'Less is More' approach in their data center architecture by consolidating core networking, load balancing, and observability into Cilium and how Alibaba is using Tetragon to secure AI agents beyond just prompt guardrails. These stories will demonstrate how Cilium is streamlining operations and reshaping the cloud native stack, cementing Cilium’s role as the networking and security data plane for modern infrastructure for the next decade to come.", "startsAt": "2026-09-08T15:30:00", "endsAt": "2026-09-08T16:00:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "25cc1809-2d06-4971-a2c9-bfdfa4538d5b", "b244f2b8-cb23-4be6-948d-58010ee7e019", "db82288a-71ae-4543-83df-49d5fc57227b" ], "categoryItems": [ 485108, 447045, 486108 ], "questionAnswers": [], "roomId": 85688, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1180062", "title": "Self-Healing Rollouts: Automating Production Fixes with Agentic AI", "description": "Your software rollouts to production are probably always flawless, right? For the rest of us, even with robust CI/CD we do run into occasional production rollout issues. In Kubernetes, Argo Rollouts excels at Progressive Delivery and automated rollbacks, but what if we could go a step further?\r\n\r\nThis session explores how to elevate your release process by integrating Agentic AI and asynchronous coding agents with canary deployments. We'll demonstrate how an intelligent agent can automatically analyze a rollout failure, pinpointing the root cause. Beyond diagnosis, these agents can take proactive steps on your behalf, suggesting and even implementing code fixes as new PRs, which can be redeployed automatically after PR review. This approach moves us closer to truly self-healing deployments.\r\n\r\nJoin us to learn how to combine the power of Kubernetes and Argo Rollouts with the autonomous capabilities of Agentic AI, achieving a release experience that is not only seamless but also resilient.", "startsAt": "2026-09-08T16:15:00", "endsAt": "2026-09-08T16:45:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "a4305a38-d462-482b-9e4a-536eb74ec63c", "4b82436a-947c-45bb-9d5d-3b94c02c8faa" ], "categoryItems": [ 447049, 446826, 447044 ], "questionAnswers": [ { "questionId": 134032, "answerValue": "https://sessionize.com/download/lrafcew~phbV87CDwzi9wMN69awkT2.pdf~kubecon-china-self-healing-rollouts_-automating-production-fixes-with-agentic-ai.pdf" }, { "questionId": 134033, "answerValue": "https://github.com/kdubois/progressive-delivery" }, { "questionId": 134034, "answerValue": "https://github.com/kdubois/kubernetes-aiops-agent" }, { "questionId": 134035, "answerValue": "https://github.com/kdubois/argo-rollouts-quarkus-demo" }, { "questionId": 134036, "answerValue": "https://github.com/argoproj-labs/rollouts-plugin-metric-ai" } ], "roomId": 80134, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1224457", "title": "Cybertwin-based Cloud Native Network (CCNN): Network Architecture Innovation and Practice", "description": "Driven by IoT and AI, future networks demand an architectural revolution. A CCNN is proposed that converges RAN, IP bearer and data center networks, using Kubernetes as the network OS to unify virtualization, scheduling, and management of compute, storage and network resources, thus enabling secure, cross-domain resource integration. The demo comprises: a Cybertwin-based personal agent service; the security agent function of Cybertwin that implements zero-trust; the data agent function that provides secure multi-dimensional personal data services; the communication agent function that delivers a cloud-native wireless access method for users; finally, a distributed medical collaborative inference based on the user’s personal agent, which demonstrates application-centric unified scheduling and intelligent orchestration of transmission, computing, and storage resources, laying an original, cutting-edge foundation for resource networking and intelligent scheduling in cloud-native networks.", "startsAt": "2026-09-08T16:15:00", "endsAt": "2026-09-08T16:45:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "8f36d40e-c5d9-45fb-9982-88829e7eab04", "0b3bfa50-14ae-4363-bf39-d130fe277c1e" ], "categoryItems": [ 447054, 446824, 447044 ], "questionAnswers": [], "roomId": 80135, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1221033", "title": "Zuul and OpenClaw, Safe Agentic coding via Human Collaboration Tools", "description": "OpenClaw isn't normally used side by side with \"safe\". But it turns out, for the past 15 years OpenDev and Zuul have been building tools and processes to support safe collaboration between thousands of unknown parties. These tools turn out to be AMAZING at unleashing the potential of systems like OpenClaw, all you have to do is treat your AI agents like untrusted people. What could possibly go wrong?\r\n\r\nUsing work on the WanderTracks project as an example, we'll walk through the architecture and approach, showing examples of separate long-lived coding and review agents interacting via the same Gerrit review and Zuul gating system we use for everyone else. We'll look at the gerrit and zuul plugins for OpenClaw that allow using code review as communication channels, and we'll highlight the use of Matrix as the primary human to agent communication channel, just like how the Zuul and OpenDev projects. And we'll highlight on the way we're running OpenClaw itself (note, not on my laptop)", "startsAt": "2026-09-08T16:15:00", "endsAt": "2026-09-08T16:45:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "d5414f6a-5825-45e1-aac1-801ff7ad9baa" ], "categoryItems": [ 447049, 446826, 447044 ], "questionAnswers": [], "roomId": 80136, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1215557", "title": "⚡ Can Stateful AI Agents Scale to Zero?", "description": "AI agents are often treated as always-on workloads because they keep session context, invoke tools, and may depend on sandboxed runtimes. Knative, by contrast, is designed for elastic, scale-to-zero services. This talk shows how the two can work together: by externalizing agent state, separating session continuity from execution, and pooling heavy runtimes, stateful agents can still benefit from serverless elasticity. We will also cover where this model works well—and where always-on runtimes remain the better choice.", "startsAt": "2026-09-08T16:15:00", "endsAt": "2026-09-08T16:20:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "aaac253c-cd23-4c13-acfb-f65c76175bc6" ], "categoryItems": [ 485112, 446827, 447045, 485120 ], "questionAnswers": [ { "questionId": 134032, "answerValue": "https://sessionize.com/download/inyascer~rnb26urcM9gvT96VtXLRWJ.pdf~kubecon-can-stateful-ai-agents-scale-to-zero.pdf" } ], "roomId": 80137, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1219807", "title": "Evolving Argo Workflows for AI: New Features, Tuning, and Multi-Tenant Isolation", "description": "Over the past year, Argo Workflows adoption in AI has surged, particularly in China. Key adopters include BYD, which runs autonomous driving pipelines on Argo, and Qwen3-Coder-Next, which models agentic coding tasks as Argo workflows in its technical report.\r\n\r\nAs engineers responsible for scaling Argo Workflows in these production environments, we share our insights from operating it as the control plane for large-scale AI pipelines.\r\n\r\nWe will cover:\r\n• Argo Workflows in Emerging Scenarios: Mapping core capabilities and execution models to autonomous driving, LLM fine-tuning, and agentic coding.\r\n• New Features and Tuning: Recent community enhancements for large-scale workloads, and practical controller tuning patterns for concurrency and queue management.\r\n• Data Isolation and Secure Access: Managing multi-tenant pipelines with per-replica, object-storage-backed volumes, and workflow-aware storage policies so each branch, step, or agent accesses only its own data subset.", "startsAt": "2026-09-08T16:15:00", "endsAt": "2026-09-08T16:45:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "975d742e-38fd-4b5f-a244-90b146c587a3", "db82288a-71ae-4543-83df-49d5fc57227b" ], "categoryItems": [ 485108, 447045, 486114 ], "questionAnswers": [], "roomId": 85688, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1212199", "title": "⚡Edge Resource Control in KubeEdge:Safely Holding and Releasing Pod Updates for Critical Edge Device", "description": "In safety-critical edge scenarios like drones in flight, autonomous robots, or AGVs/AMRs, an unplanned pod restart can cause motion interruption, safety hazards, or crashes. Traditional Kubernetes assumes updates can happen anytime, but many edge applications need explicit confirmation that the device is safe (landed, parked, or idle) before upgrading.This lightning talk introduces Edge Resource Upgrade Control, a new feature in KubeEdge v1.22.0. By adding the annotation edge.kubeedge.io/hold-upgrade: \"true\" to Deployments, StatefulSets, DaemonSets, or Pods, the edge node intercepts the new Pod and holds it in a pending state inside edged. A new HeldUpgrade PodCondition is reported back to the cloud.On the edge, local logic decides when it is safe. Operators release the hold with:keadm ctl unhold-upgrade pod (pod-name)\r\nkeadm ctl unhold-upgrade node (node-wide)", "startsAt": "2026-09-08T16:22:00", "endsAt": "2026-09-08T16:27:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "93d37137-a3ac-42e1-b31c-41d4fbf86253" ], "categoryItems": [ 485112, 446826, 447044, 485116 ], "questionAnswers": [ { "questionId": 134032, "answerValue": "https://sessionize.com/download/ilpavdek~RPVGeztevtqCMdAMGHQwkQ.pptx~edgeresourcecontrolinkubeedge.pptx" } ], "roomId": 80137, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1222625", "title": "⚡ Eliminating Adaptation Latency: Day-0 Inference for New Models on Multi-Backends with vLLM", "description": "As LLM iterations accelerate, a significant \"adaptation latency\"—the days or weeks required to port models from GPU to heterogeneous hardware delays product launches. This bottleneck stems from the lack of clear abstraction boundaries between inference frameworks and hardware backends, forcing invasive code modifications for each new model. \r\nThis session decodes how vLLM compresses Ascend NPU adaptation from \"per-model porting\" to Day-0 Inference through modular design. This session will demonstrate: 1) How pluggable architecture and Out-of-Tree adaptation allow independent development without main-branch blocking; 2) How Custom OP mechanisms enable non-invasive, generalized operator extensions; and 3) How main2main synchronization and robust CI systems ensure Ascend backends stay aligned with vLLM's rapid releases. These capabilities reduce adaptation cycles from weeks to hours and provide a reproducible paradigm for rapid integration within the AI hardware ecosystem.", "startsAt": "2026-09-08T16:29:00", "endsAt": "2026-09-08T16:34:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "305b46aa-7ea7-4056-8f3f-8ca103db88c8" ], "categoryItems": [ 485112, 446827, 447045, 485118 ], "questionAnswers": [], "roomId": 80137, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1184985", "title": "⚡Empower open edge computing on RISC-V", "description": "LF Edge - Linux Foundation Edge, is aiming to establish an open, interoperable software framework for edge computing, independent of hardware, silicon, cloud, but limited on x86 and arm. Meanwhile, open hardware like RISC-V are gaining ground in IoT and embedded systems, with furthering the expansion of edge computing. We believe the blend of such a open edge software, open RISC-V hardware with open LF Edge can herald a new era of empowering Open Software on Open Hardware, to unlock the power of edge computing everywhere. \r\n\r\nHe we'd like to review how we enable-to-build modern edge platform on real RISC-V hardware platform in the real world by walking through current state, benefits, challenges, and our solutions. Especially, we will demonstrate this with great demo by enabling LF Edge project like EdgeX and edge AI to RISC-V Linux and further show this on heterogeneous hardware platforms along with Linux by deploying on multiple arches as needed.", "startsAt": "2026-09-08T16:36:00", "endsAt": "2026-09-08T16:41:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "42620d7f-2a95-4c85-917b-6baf8e290916" ], "categoryItems": [ 485112, 446826, 447045, 485116 ], "questionAnswers": [], "roomId": 80137, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": false }, { "id": "1193509", "title": "⚡ Fast Restarts, Not Just Fast Starts: Accelerating Pod Recovery", "description": "Pod startup optimization is no longer enough; restart latency is now a core bottleneck for reliability and cost, especially in AI/ML and batch workloads. This session connects start-up and restart as one lifecycle problem. We begin with a practical taxonomy (container restart vs Pod recreation vs rollout/eviction), then break down where restart time is lost: re-scheduling, image pull, init re-execution, and control-plane churn.\r\n\r\nThe main case study is JobSet in-place restart, where prototype results show recovery improving from 2m10s to 10s at 5,000-node scale. We map this to new Kubernetes capabilities such as per-container restart policies/rules and restart-all-containers. Finally, we provide a production checklist across kubelet tuning, workload controller design, probes/signals, and observability so teams can safely reduce restart time without harming SLOs.", "startsAt": "2026-09-08T16:43:00", "endsAt": "2026-09-08T16:48:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "e355bcd9-b499-4179-8a69-5d6a850b4d2e" ], "categoryItems": [ 485112, 446827, 447045, 485114 ], "questionAnswers": [ { "questionId": 134032, "answerValue": "https://sessionize.com/download/mvaotte~aXUZycoNMCq6HJsAK45S4o.pptx~fast-restarts.pptx" } ], "roomId": 80137, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1224946", "title": "⚡ Speedy But Sturdy PyTorch Optimizers", "description": "PyTorch strives to offer tested and proven optimizers (think SGD, AdamW, Muon) and we want them to be fast, consistent, and simple to use. We want people to trust our optimizers :D This session will cover how we pushed along each of those dimensions with our latest developments on newer optimizers like Muon, adding more mixed precision (BF16 AdamW), and lifting out-of-the-box performance. We’ll share the new features and algorithms you can try out for driving down your loss curves and, if you’re curious, give insight on how we make things speedier but keep them just as sturdy.", "startsAt": "2026-09-08T16:50:00", "endsAt": "2026-09-08T16:55:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "bcfc2c5d-dca4-47dc-affa-62de55829f5e" ], "categoryItems": [ 485112, 446827, 447044, 485123 ], "questionAnswers": [ { "questionId": 134032, "answerValue": "https://sessionize.com/download/ttavnew~KeEKMYdVrUMoDuJANVUuYH.pdf~ptc-china-2026-optimizers.pdf" } ], "roomId": 80137, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1225153", "title": "⚡Building a Multi-Cluster Progressive Delivery Platform with Karmada & Argo", "description": "As enterprises scale across hybrid clouds, the gap between \"cluster scheduling\" and \"release control\" remains a significant hurdle. While many understand why Karmada and Argo are a powerful duo, the technical \"how\" of integrating them into a seamless, global delivery pipeline is often left to the imagination.\r\nIn this demo-heavy session, we move beyond the theory to show you exactly how to build a unified multi-cluster progressive delivery platform. We will live-demonstrate a complete canary release workflow—from initial configuration to global status aggregation—driven by a single declarative specification.\r\nKey takeaways include:\r\n1. Deep Integration Architecture: How to run ArgoCD and Argo Rollouts with Karmada.\r\n2. Resource Interpretation: Configuring Karmada’s Resource Interpreter to aggregate ReplicaSet status for global observability.\r\n3. Implementation Details: A step-by-step breakdown of the installation, CRD configurations, and propagation policies.", "startsAt": "2026-09-08T16:57:00", "endsAt": "2026-09-08T17:02:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "d3d7d2b1-05c9-4494-be6f-9a2d13eef4e8" ], "categoryItems": [ 485112, 446827, 447044, 485120 ], "questionAnswers": [], "roomId": 80137, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1222276", "title": "Agile Readiness: Leveraging CI Relay and Full-Scale Test Reuse", "description": "Adapting to PyTorch’s rapid bi-monthly release cycle is a challenge for heterogeneous accelerators. This session shares Ascend’s practices in maintaining high quality while achieving stable releases within one month of upstream updates.\r\n\r\nTest Refactoring & Device-Agnostic Reuse:Learn how we use instantiate_device_type_tests and dynamic skipping to decouple tests from hardware. This allows seamless reuse of 580K+ existing community test cases, ensuring full API semantic parity and feature validation at minimal cost.\r\n\r\nCross-Repo CI Relay (CRCR) & Quality Gates:We introduce the CRCR mechanism, enabling real-time co-compilation and testing between PyTorch PRs and accelerator code. By intercepting regressions during the PR stage, we mitigate version adaptation pressure and ensure rapid post-release stability.\r\n\r\nThrough automated test reuse and cross-repo CI, Ascend NPU delivers high-quality releases within 30 days, significantly enhancing user experience.", "startsAt": "2026-09-08T17:00:00", "endsAt": "2026-09-08T17:30:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "66e515d5-3e82-4f74-864e-5dc52a0a7985", "71d5f45d-6d99-4e32-a0fc-54bc07b69532" ], "categoryItems": [ 447056, 446827, 447045 ], "questionAnswers": [], "roomId": 80134, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1210983", "title": "Gravity of Ecosystems: Pollinating the Modern AI Orchestrator", "description": "Open source isn’t just code; it’s the human gravity that keeps our industry moving. However, for too long, infrastructure, orchestration, and AI frameworks have behaved like oil and water—occupying the same space but never truly mixing. This friction creates massive inefficiencies, from \"double scheduling\" gaps to workloads that remain blind to the hardware they run on.\r\n\r\nIn this session, Antonio Ojea and Wenjia Zhang share how we are breaking these silos through active pollination. This is not a mere declaration of intent; it is the result of over a year of moving core contributors into projects like Ray, Slurm, and PyTorch and starting dedicated initiatives within Kubernetes. We’ll explore how this hands-on work is evolving the K8s core through Dynamic Resource Allocation (DRA) and the Workload API. By collaborating across foundations instead of working in isolation, we achieve the efficiency and community sovereignty needed to keep the future of AI open and in our hands.", "startsAt": "2026-09-08T17:00:00", "endsAt": "2026-09-08T17:30:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "83fc1a01-1dc2-438c-8e99-b036ba9b67be", "93496157-eb91-4ced-8ff2-51d506e10493" ], "categoryItems": [ 447054, 446827, 447045 ], "questionAnswers": [], "roomId": 80135, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": false }, { "id": "1224575", "title": "Redefining LLM Training through Atomic Components and Serverless TaaS", "description": "RL (e.g., GRPO) is vital for LLM reasoning, yet current tools are either too rigid or engineering-heavy. Twinkle, co-developed by ModelScope and China Merchants Bank, offers a breakthrough via:\r\n- Atomic Design: Orchestrate RL loops in ~150 lines. Swap modules (Reward/Loss) like Lego without touching the core.\r\n- C-S Architecture: Decouples algorithm semantics from GPU orchestration (torchrun/Ray/HTTP).\r\n- Serverless TaaS: Multi-tenant LoRA pools enable 8+ concurrent tasks on one base model, maximizing resource utility.\r\nAs a superset of the Tinker API, Twinkle supports FSDP2 and Megatron backends. It solves the flexibility-scale trade-off, enabling true \"Training-as-a-Service.\" We will explore its architecture, rapid GRPO/DPO integration, and financial sector case studies, concluding with the shift toward \"micro-service training\" for Agent RL.", "startsAt": "2026-09-08T17:00:00", "endsAt": "2026-09-08T17:30:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "5bccdb2d-dac4-4b1f-a0b5-a7355828e71d" ], "categoryItems": [ 447046, 446827, 447045 ], "questionAnswers": [ { "questionId": 134032, "answerValue": "https://sessionize.com/download/ciwcacre~GLfFPhvYZigaytCzfBDrC8.pdf~twinkle_deck.pdf" } ], "roomId": 80136, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1224605", "title": "Hibernate and Evolving: Taming OpenClaw with OpenKruise Agents", "description": "OpenClaw's rapid adoption has highlighted critical \"Day-2\" operational challenges: high costs of idle agents, sandbox instability, and complex version upgrades. This session presents a production-ready best practice using OpenKruise Agents to manage these agents efficiently on Kubernetes.\r\n\r\nWe will demonstrate three core strategies:\r\n1. Secure Execution: Enforcing network isolation and secure API key injection \r\n2. Cost Optimization: Implementing automatic hibernation to scale resources of idle agents to zero, slashing infrastructure costs.\r\n3. Zero-Downtime Upgrades: Evolving OpenClaw versions seamlessly without losing agent state or data.\r\nFinally, we will discuss integrating the upcoming Kubernetes Checkpoint/Restore API to enable vendor-neutral sandbox hibernation. Join us to transform your OpenClaw deployment into a scalable, cost-efficient system.\r\n", "startsAt": "2026-09-08T17:00:00", "endsAt": "2026-09-08T17:30:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "a6e8b69b-7ea1-4686-a849-52d913553d69" ], "categoryItems": [ 485108, 447044, 486109 ], "questionAnswers": [], "roomId": 85688, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1220735", "title": "⚡ Before vLLM starts: Preflight Checks for LWS for LLM Inference on K8S", "description": "Training usually run preflight tests(NCCL, topology, Connectivity) to avoid discovering infra problems after job running. \r\nInference needs the same discipline. In distributed serving, the most expensive failures are often not startup failures, but clusters that come up and then show unstable throughput, p99 latency jitter, or degraded cross-node performance, with high debugging cost later.\r\n\r\nWe presents a concrete LWS-based startup gate for distributed inference: run preflight in init-containers, fail early when invariants are broken, and keep checks decoupled from vLLM. As my LWS KEP #813 as the case study (https://github.com/kubernetes-sigs/lws/pull/813), we focus on 3 gaps hit in production: init-phase peer discovery, unbounded recreate loops, and unclear terminal failure boundaries. \r\n\r\nTo make it practical, we will also share useful examples of preflight checklist, and how to map each failure class to retry, reschedule, or stop.\r\n\r\nFail Early, Inference Better :-)\r\n", "startsAt": "2026-09-08T17:04:00", "endsAt": "2026-09-08T17:09:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "aa1d25a7-7dfe-4fa4-ae67-59c835e5dbae" ], "categoryItems": [ 485112, 446826, 447044, 485115 ], "questionAnswers": [ { "questionId": 134032, "answerValue": "https://sessionize.com/download/ipcabzec~pCjHasFF5EvMw5K29jsEgj.pptx~kubecon-2026-peter-preflight-and-lws.pptx" } ], "roomId": 80137, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1255206", "title": "Welcome Reception 🎉", "description": "Join us onsite for drinks, appetizers, and conversations with old and new friends in the Solutions Showcase. Explore the exhibit booths to learn more about the latest technologies, meet experts and project maintainers, browse special offers, and much more.\r\n\r\nIn order to facilitate networking and business relationships at the event, you may choose to visit a third party’s booth or to access sponsored content. You are never required to visit third-party booths or to access sponsored content. When visiting a booth or participating in sponsored activities, the third party will receive some of your registration data. This data includes your first name, last name, title, company, address, email, standard demographics questions (i.e. job function, industry), and details about the sponsored content or resources you interacted with. If you choose to interact with a booth or access sponsored content, you are explicitly consenting to receipt and use of such data by the third-party recipients, which will be subject to their own privacy policies.", "startsAt": "2026-09-08T17:30:00", "endsAt": "2026-09-08T19:00:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [], "categoryItems": [ 485111 ], "questionAnswers": [], "roomId": 85878, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": false }, { "id": "1255198", "title": "Registration + Badge Pick-Up", "description": null, "startsAt": "2026-09-09T08:00:00", "endsAt": "2026-09-09T16:00:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [], "categoryItems": [ 485107 ], "questionAnswers": [], "roomId": 85877, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": false }, { "id": "1255202", "title": "Cloakroom", "description": null, "startsAt": "2026-09-09T08:00:00", "endsAt": "2026-09-09T17:30:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [], "categoryItems": [ 485107 ], "questionAnswers": [], "roomId": 88309, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": false }, { "id": "1304839", "title": "Welcome Back + Opening Remarks", "description": null, "startsAt": "2026-09-09T09:00:00", "endsAt": "2026-09-09T09:05:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "af7b6fb1-4084-4706-a495-3fc1793d97f2", "a50dbbe3-5e92-44ca-9e40-c1b39986cbc7" ], "categoryItems": [ 482870, 447044 ], "questionAnswers": [], "roomId": 80136, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1312069", "title": "What Powers Frontier Intelligence", "description": "Models are becoming more capable, inference more efficient, and cloud native technologies are making AI easier to operate at scale. But everything starts at the infrastructure layer. Every model, inference service and AI agent ultimately depends on the compute, accelerators, networking and storage underneath it.\r\n\r\nIf you want to build and run the AI stack in open source, the infrastructure layer has to be open too. This keynote explores how OpenInfra makes that vision real—turning heterogeneous hardware and scarce compute into infrastructure that can power AI at scale while providing the isolation, flexibility and control organizations need. From accelerated computing and trusted execution to sovereignty and infrastructure control, we'll look at how open infrastructure connects with cloud native and AI technologies to create a complete open source stack for frontier intelligence.", "startsAt": "2026-09-09T09:05:00", "endsAt": "2026-09-09T09:15:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "66f55a66-d9bd-4877-8c69-4a2186b74bef" ], "categoryItems": [ 482870, 447044 ], "questionAnswers": [], "roomId": 80136, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1304828", "title": "Road from Kata Containers to Confidential Containers + GPUs: From First Commit to CNCF Incubation", "description": "Every road has a first step. This one began in November 2021 with a single issue filed against Kata Containers, followed shortly by a first pull request. Talk by talk and release by release, that work led to Confidential Containers becoming a CNCF incubating project with GPU workloads running inside the trust boundary.\r\n\r\nThis keynote retraces the road through the milestones that shaped it: isolation through lightweight VMs, attestation that reaches from silicon to service, Kubernetes as the platform for AI/ML on private data, and confidential computing on GPUs at scale. Each stop taught a lesson.\r\n\r\nTogether, they explain where the project stands today and why. The talk ends at incubation, the current end of the road, and closes with a live demo of confidential containers with GPUs.", "startsAt": "2026-09-09T09:17:00", "endsAt": "2026-09-09T09:22:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "4d2a68f8-5859-49d1-9caa-e488d11baa9b" ], "categoryItems": [ 482870, 447044 ], "questionAnswers": [], "roomId": 80136, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1313956", "title": "HyperParallel: A SuperPoD-Aware Distributed Acceleration Library", "description": "HyperParallel is a PyTorch-based distributed acceleration library optimized for Ascend SuperPoD. This demonstration is conducted on a Huawei Atlas 800T cluster using a Qwen3-30B-A3B training workload. Under identical configurations, HyperParallel FSDP2 + Muon delivers substantially higher training throughput than PyTorch FSDP2 + Muon while maintaining consistent loss convergence. The live demonstration presents the complete distributed model training workflow and dynamically visualizes training throughput and loss convergence curves, providing an intuitive illustration of HyperParallel’s acceleration capabilities for large-model training.", "startsAt": "2026-09-09T09:24:00", "endsAt": "2026-09-09T09:29:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "74274563-3c5e-4b6b-9da6-0ab62d49fab4", "c2ed46cc-5d32-4933-a395-18af73da5d4f" ], "categoryItems": [ 482870 ], "questionAnswers": [], "roomId": 80136, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1309268", "title": "Build a Unified Heterogeneous AI Computing Ecosystem for PyTorch", "description": "AI computing hardware faces severe heterogeneity challenges with high adaptation costs and lack of unified standards. Co-chaired by Huawei and Intel, PyTorch TAC Accelerator Integration WG delivers standardized hardware onboarding guidelines, cross-repo CI testing mechanism, generalized device-aware test suites and platform incubation workflows. This keynote shares our completed achievements and future roadmap including refined device-agnostic APIs and expanded multi-backend test matrix. We invite all hardware vendors and developers to co-build an open, efficient heterogeneous computing ecosystem for PyTorch.", "startsAt": "2026-09-09T09:31:00", "endsAt": "2026-09-09T09:36:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "c20c52c1-5413-485d-acc8-de4412276233" ], "categoryItems": [ 482870, 447045 ], "questionAnswers": [], "roomId": 80136, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1317079", "title": "Serving Qwen at Scale: Multi-Cluster AI Infrastructure on Karmada", "description": "Serving Qwen requires infrastructure at massive scale: training and serving large language models across multiple Kubernetes clusters and vast GPU resources spanning multiple regions. In this keynote, the Qwen infrastructure team shares how they manage this complexity in production — unifying fragmented GPU resources, scheduling AI workloads across clusters, and keeping the platform reliable as the business grows rapidly. The talk walks through the architecture behind Qwen, the hard lessons learned along the way, and how cloud native multi-cluster technologies, including Karmada, help turn a fleet of clusters into one coherent AI platform.", "startsAt": "2026-09-09T09:38:00", "endsAt": "2026-09-09T09:41:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "7805aad6-9d9c-4ade-b72f-cc935d18679d" ], "categoryItems": [ 482870, 447045 ], "questionAnswers": [], "roomId": 80136, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1309304", "title": "Meet the Community Behind the Open Source AI Stack", "description": "No single project, company, or community builds the AI stack alone. Behind the models, frameworks, cloud native technologies, and infrastructure powering AI is a global open source community working across every layer. This session brings together contributors from across the stack to share how collaboration between projects and communities turns individual technologies into an open, interoperable AI ecosystem—and why that collaboration matters even more as AI evolves.", "startsAt": "2026-09-09T09:45:00", "endsAt": "2026-09-09T09:53:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "44e11070-13f7-47e1-8ab3-9b2a5ff64d36", "87a1caef-272a-4dd5-9390-cd7e06480345", "c20c52c1-5413-485d-acc8-de4412276233", "1351ad0e-c69b-412a-9a35-3505e6398656" ], "categoryItems": [ 482870, 447045 ], "questionAnswers": [], "roomId": 80136, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1325851", "title": "Building Frontier AI Infra: SGLang and Miles", "description": "A model’s real-world performance gets decided in the serving and post-training layer: how fast it runs, what it costs, and how quickly people can build on it. That’s the layer SGLang and Miles have been pushing hard on, especially as agentic workloads bring new challenges in cache reuse, long-context serving, and large-scale RL. In this talk, we’ll cover key features designed for agentic workloads, including Unified Radix Cache, HiCache, parallelism strategies for long-context serving, and HiSparse, as well as efficient and scalable RL with Miles, built on top of SGLang for high-performance rollouts and fully asynchronous training.", "startsAt": "2026-09-09T09:55:00", "endsAt": "2026-09-09T09:58:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "ff78d921-13c1-4c10-822d-c81dd4168e22" ], "categoryItems": [ 482870 ], "questionAnswers": [], "roomId": 80136, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1308448", "title": "A Cloud Native Stack from Bare Metal to Tokens for Large-Scale AI Inference", "description": "We didn't buy all our GPUs/AI chips at once. Years of procurement across different budget cycles left us with a fleet spanning multiple generations-from T4s that still run fine, to L40Ss, A100s, H200s, and the newer B200s-each with different memory capacities, compute profiles, and driver requirements. Managing these as isolated, team-specific bare-metal allocations meant we were sitting on underutilized assets while other teams waited in queue.\r\n\r\nThis talk is about how we unified that messy reality into a single platform by layering three open-source ecosystems: OIF at the bare-metal layer, CNCF tooling for GPU pooling and multi-tenant slicing, and PyTorch Foundation projects for serving optimization.\r\n\r\nThe result is a Token-as-a-Service platform that treats every GPU in the fleet (old or new) as a productive contributor. We'll share what it took to get there, what we're still figuring out, and where we see this going as model complexity and multi-agent workloads continue to grow.", "startsAt": "2026-09-09T10:00:00", "endsAt": "2026-09-09T10:05:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "0f17947b-02a6-404d-8878-3fe915a9639e" ], "categoryItems": [ 482870, 447044 ], "questionAnswers": [ { "questionId": 134032, "answerValue": "https://sessionize.com/download/iwvabbeb~JtLQi6m8kTTM9XEAGaKDgz.pdf~viettel-ai-platform-2026.pdf" } ], "roomId": 80136, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1304845", "title": "Closing Remarks", "description": null, "startsAt": "2026-09-09T10:25:00", "endsAt": "2026-09-09T10:30:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [], "categoryItems": [ 482870, 447044 ], "questionAnswers": [], "roomId": 80136, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1255190", "title": "Coffee Break ☕ ", "description": null, "startsAt": "2026-09-09T10:30:00", "endsAt": "2026-09-09T11:00:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [], "categoryItems": [ 485106 ], "questionAnswers": [], "roomId": 85878, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": false }, { "id": "1311158", "title": "Project Pavilion | Wednesday AM Project Tables ", "description": "Wednesday AM Project Tables | 10:30 - 12:45\r\n\r\nT-1\tOpenEverest\r\nT-2\tKarmada\r\nT-3 Cilium\r\nT-4\tDragonfly\r\nT-5\tCNCF Table\r\nT-6\tCommunity-Led Events\r\nT-7\tllm-d\r\nT-8\tspiderpool\r\nT-9\tFluid\r\nT-10 hwameistor\r\nT-11 Kata Containers (OpenInfra project)\r\nT-12 DeepSpeed (PyTorch project)\r\nT-13 kubean\r\n\r\n\r\nView the full project table directory here: https://www.lfopensource.cn/kubecon-cloudnativecon-openinfra-summit-pytorch-conference-china/features-add-ons/project-engagement/#project-table-directory", "startsAt": "2026-09-09T10:30:00", "endsAt": "2026-09-09T12:45:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [], "categoryItems": [ 485109 ], "questionAnswers": [], "roomId": 85878, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1255195", "title": "Solutions Showcase", "description": null, "startsAt": "2026-09-09T10:30:00", "endsAt": "2026-09-09T15:30:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [], "categoryItems": [ 485110 ], "questionAnswers": [], "roomId": 85878, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": false }, { "id": "1224671", "title": "Serving AI at Massive Scale: The Cloud-Native Inference Plane Behind Huawei Celia", "description": "As AI shifts to integrated mobile assistants, serving multimodal AI across a massive device ecosystem demands managing extreme traffic and strict latency. Prefill-Decode (PD) disaggregation is the essential paradigm, but its sharp compute and bandwidth asymmetry fundamentally breaks traditional Kubernetes Workloads, Routers, and HPA.\r\n\r\nExperts from Huawei Consumer Business and Kthena will publicly unpack the infrastructure powering Huawei Celia — one of the world's largest mobile AI assistants. Backed by tens of thousands of accelerators, the presentation will detail PD-disaggregated serving at scale.\r\n\r\nLeveraging PodGroup and SubGroup gang scheduling, speakers will explore orchestrating the Inference Plane across Workload, Router, and AutoScaler. The session will analyze real traffic deltas (−40% TTFT, +50% throughput, utilization, and cost-per-million-tokens). Attendees will leave knowing how to model PD workloads, route on tokens, and scale Prefill and Decode using the right metrics.", "startsAt": "2026-09-09T11:00:00", "endsAt": "2026-09-09T11:30:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "1b814ba0-bc81-4245-880a-097b1e7e3b75" ], "categoryItems": [ 447047, 446826, 447045 ], "questionAnswers": [], "roomId": 80134, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1224844", "title": "Pathless First: Redefining runc Container Security Against ProcFS-Based Attacks", "description": "ProcFS has been the most persistent attack surface for runc container escapes, accounting for 60% of path-related runc CVEs from 2017 to 2025. Incremental patches and path hardening have repeatedly failed to address the root cause: runc’s reliance on attacker-controllable ambient path resolution.\r\nAs a runc maintainer and lead researcher of the Pathless Container Setup (PCS) framework, I will trace the evolution of ProcFS-based attacks from early TOCTOU symlink hijacking to modern bind mount escapes. I will then present the PCS framework’s file descriptor-centric design, which leverages modern Linux pathless mount APIs (fsopen, fsmount) to completely eliminate path resolution risks. This design has been upstreamed into runc and deployed across global public clouds.\r\nAttendees will learn how this approach mitigates all 9 known Path-related CVEs and gain actionable best practices to harden their own container runtime deployments.", "startsAt": "2026-09-09T11:00:00", "endsAt": "2026-09-09T11:30:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "b6340cc0-5435-4d62-8352-7bda98552c72" ], "categoryItems": [ 447051, 446826, 447045 ], "questionAnswers": [ { "questionId": 134032, "answerValue": "https://sessionize.com/download/powuy~ReFLDezCLCUwpTkPzRqtMo.pdf~kubecon2026-pathless-first.pdf" }, { "questionId": 134033, "answerValue": "https://github.com/opencontainers/runc" }, { "questionId": 134034, "answerValue": "https://lifubang.github.io/acm/" }, { "questionId": 134036, "answerValue": "https://lifubang.github.io/acm/golang-new-mount-api/" } ], "roomId": 80135, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1219415", "title": "Escaping the Vendor Trap: A Journey for Migrating Legacy Infrastructure to OpenStack and K8s", "description": "In 2018, most of IT and Telco infras of Viettel (largest telco in Vietnam) run on VMware or legacy on-premise solutions. The utilization of physical infra was low, need hundreds of people to operate all the infras, MTTR was nightmare, depends on support tickets from vendors... Recently, VMWare licenses price went high. That's why we have to migrate our infras to open source tech stacks and solutions. OpenStak, Ceph, Kubernetes, Prometheus... are our choices. By the end of 2025, over 80% of our massive infra had been migrated to open source Cloud and Cloud Native solutions.\r\nThis talk will be about how we choose the starting point, how we move from here to there, and the way we achieve the current state, how we collaborated with open source community in Vietnam and Asia.", "startsAt": "2026-09-09T11:00:00", "endsAt": "2026-09-09T11:30:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "0f17947b-02a6-404d-8878-3fe915a9639e", "03dbf5de-879a-4c7c-b9c0-46d68383c559" ], "categoryItems": [ 447053, 446826, 447044 ], "questionAnswers": [ { "questionId": 134032, "answerValue": "https://sessionize.com/download/wkalwem~3NEA7pbsSRiZT12thET6M1.pdf~viettel-escaping-vendor-trap-2026.pdf" } ], "roomId": 80136, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1224867", "title": "In-Place PVC Re-Binding: Zero-Downtime Disk Migration Using Only the Kubernetes API", "description": "Running stateful workloads in Kubernetes is challenging; migrating the storage beneath them is even harder. Migrating hundreds of live storage volumes without downtime might seem nearly impossible. In this session, we present a battle-tested approach for doing exactly that: migrating hundreds of Azure Disk-backed PersistentVolumes from deprecated in-tree drivers to CSI with zero downtime and minimal service disruption. By leveraging a deep understanding of Kubernetes internals, we accomplished this using only standard user-facing Kubernetes APIs without control plane forks, risky hacks, or even custom operators.\r\n\r\nWe’ll break down the critical mechanics that make it possible, touching upon immutability constraints, claimRef behavior, finalizers, and StorageClass interactions. Through real-world examples from large-scale production data platforms, this talk demonstrates how advanced Kubernetes systems expertise can unlock operational capabilities that many teams assume are out of reach.", "startsAt": "2026-09-09T11:00:00", "endsAt": "2026-09-09T11:30:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "8010363b-334b-482e-825d-ee6acf5c7cf2", "9566ce0e-7307-4fd9-a90f-d63a1c50131c" ], "categoryItems": [ 447053, 446826, 447044 ], "questionAnswers": [], "roomId": 80137, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1304033", "title": "Accelerating RL with AgentCube: Cloud-Native Multi-Agent Collaboration", "description": "Reinforcement learning workloads often require multiple agents to start concurrently, interact with environments, evaluate rewards, and iteratively update policies. However, Pod cold starts, runtime isolation, and lifecycle management can become bottlenecks for large-scale agent collaboration.\r\n\r\nThis demo shows how AgentCube provides Kubernetes-native runtime infrastructure for RL workflows. An RL controller creates multiple agent sessions through a unified API, while AgentCube rapidly provisions isolated Pod-based sandboxes using warm pools. The agents then collaborate to perform inference, tool invocation, code execution, and result processing before returning data for reward calculation and policy updates.\r\n\r\nThe demo will highlight fast startup, multi-agent coordination, session routing, elastic scaling, resource reuse, secure isolation, and automatic cleanup. By integrating with Kubernetes and Volcano scheduling, AgentCube enables high-concurrency, short-lived RL agent workloads with a serverless-like experience.\r\n\r\nIn order to facilitate networking and business relationships at the event, you may choose to visit a third party’s booth or access sponsored content. You are never required to visit third party booths or to access sponsored content. When visiting a booth or participating in sponsored activities, the third party will receive some of your registration data. This data includes your first name, last name, title, company, address, email, standard demographics questions (i.e. job function, industry), and details about the sponsored content or resources you interacted with. If you choose to interact with a booth or access sponsored content, you are explicitly consenting to receipt and use of such data by the third-party recipients, which will be subject to their own privacy policies. ", "startsAt": "2026-09-09T11:00:00", "endsAt": "2026-09-09T11:20:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "14ebc104-3e6c-4e50-a410-5398b6d83908" ], "categoryItems": [ 500857 ], "questionAnswers": [], "roomId": 85688, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1303982", "title": "KV Cache: Accelerating AI inference on Intel CPU", "description": "Serving LLM is bottlenecked by one scarce resource: HBM. The KV Cache grows with every token and every concurrent request, capping context length, throughput, and ultimately the number of users a cluster can serve. Teams respond by buying more GPUs —paying for compute they don't need just to get the HBM they do.\r\n\r\nThis talk introduces KVShrink, an approach that treats the KV cache as a tiered, compressible asset instead of a fixed HBM cost. KVShrink offloads the KV cache out of HBM into DDR/Storage and compresses it with Intel QAT hardware accelerator - retaining cache that would otherwise be evicted so it can be reused across requests. Higher hit rates mean the GPU skips redundant re-computation, cutting TTFT latency, improving the user experience, and sustaining far higher concurrency.\r\n\r\nIn order to facilitate networking and business relationships at the event, you may choose to visit a third party’s booth or access sponsored content. You are never required to visit third party booths or to access sponsored content. When visiting a booth or participating in sponsored activities, the third party will receive some of your registration data. This data includes your first name, last name, title, company, address, email, standard demographics questions (i.e. job function, industry), and details about the sponsored content or resources you interacted with. If you choose to interact with a booth or access sponsored content, you are explicitly consenting to receipt and use of such data by the third-party recipients, which will be subject to their own privacy policies. ", "startsAt": "2026-09-09T11:25:00", "endsAt": "2026-09-09T11:45:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "3872d9f6-ea0a-4421-a994-a803e688411f" ], "categoryItems": [ 500857, 447045 ], "questionAnswers": [ { "questionId": 134032, "answerValue": "https://sessionize.com/download/ijzappeg~GghacyHV9Jff6YjjfU19pa.pdf~intel_accelerating_ai_inference_on_intel_cpu.pdf" } ], "roomId": 85688, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1223450", "title": "Training Through Failures: How Meta Keeps 100k-GPU Jobs Alive with Open-Source Fault Tolerance", "description": "GPU failures at scale are not exceptions — they are scheduled events. At Meta, training jobs spanning tens of thousands of GPUs encounter hardware faults daily, yet training continues without missing a step. In this talk, we discuss open-source tools that make this possible.\r\n\r\nWe introduce the fault tolerance APIs in torchcomms, covering NCCL, MCCL and Gloo backends. MCCL is Meta's production fault tolerance collective communication library and we will deep dive into its design and fault tolerance capabilities. We also show how torchcomms uses NCCL's new primitives — ncclShrink, ncclGrow, and ncclCommRevoke — to gracefully shrink or grow communicator membership without deadlocking the cluster.\r\n\r\nWe'll cover novel checkpointing approaches — including checkpoint-to-peer-memory and checkpoint-to-local-disk — using torchstore as one example of how fast RDMA transfers between GPU peers can replace slow distributed filesystem writes and enable live recovery.", "startsAt": "2026-09-09T11:45:00", "endsAt": "2026-09-09T12:15:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "dde599d6-57ee-4fbc-9a07-ae71ff0c3c61", "dd6c32e8-7b29-4f89-9026-344f951594b8" ], "categoryItems": [ 447047, 446826, 447044 ], "questionAnswers": [], "roomId": 80134, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1218125", "title": "All-or-Nothing No More: Taming Kubernetes Impersonation with Granular Controls", "description": "Kubernetes impersonation lets one identity act as another, but today it's all-or-nothing — an impersonator gains the target's full privileges, often exceeding what's actually needed.\r\n KEP-5284 introduces constrained impersonation, allowing clusters to enforce least-privilege identity delegation by restricting both who can be impersonated and what actions the\r\n impersonator may perform. This unlocks safer patterns for real-world use cases: per-node agents like CSI or CNI plugins that need to impersonate a node but are restricted to only their\r\n required actions, and deputy controllers that securely proxy user operations without inheriting full privileges. Now in beta, we'll walk through the newly introduced authorization\r\n verbs, how constrained impersonation is designed to scale, adoption status from early adopters, and practical guidance for migrating from unrestricted impersonation.", "startsAt": "2026-09-09T11:45:00", "endsAt": "2026-09-09T12:15:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "6e6ea1fd-e7fb-4e69-9958-aca0438036e6" ], "categoryItems": [ 447051, 446826, 447045 ], "questionAnswers": [ { "questionId": 134032, "answerValue": "https://sessionize.com/download/oksagu~sMh8cviXybcy4WZEN3s9K4.pdf~impersonation-talk.pdf" } ], "roomId": 80135, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1224832", "title": "Secure AI Agent Sandboxing with OpenStack Zun and Kata Container", "description": "AI Agents are moving to production, but executing LLM-generated code securely at scale remains a critical gap for OpenStack-native environments. Rather than bolting on external services or accepting inadequate isolation, this session presents a secure sandbox built natively on OpenStack.\r\nWe’ll share how we combined Zun’s orchestration agility with Kata’s hardware-level isolation to run hundreds of ephemeral, untrusted execution environments, and how this security layer integrates with our broader Agent infrastructure for heterogeneous GPU scheduling and dynamic skill discovery.\r\nThis talk will walk through our end-to-end flow: how we leverage OpenStack Glance for sandbox image management, Neutron for isolated tenant networking, and Zun for container lifecycle orchestration; how Kata’s lightweight virtualization provides the hardware-level security boundary between untrusted agent code and the host.", "startsAt": "2026-09-09T11:45:00", "endsAt": "2026-09-09T12:15:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "32e5b704-5b3e-4a4c-ac27-9cbcbd6da7a9" ], "categoryItems": [ 447046, 446826, 447044 ], "questionAnswers": [ { "questionId": 134032, "answerValue": "https://sessionize.com/download/idbatkeq~wNbLV3hoKvGDdd9EL11Fiq.pdf~secure-ai-agent-sandbox-upload.pdf" } ], "roomId": 80136, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1221758", "title": "One Size Fits None: Lightweight Workload-Specialized Confidential Containers", "description": "Confidential Containers (CoCo) brought seamless data-in-use protection to K8s via a Pod-as-a-CVM model with in-guest image pulling and attestation. However, this \"compatibility-first\" approach comes at the cost of increased complexity, bloated guests, and slow boot times - undermining previous efforts to make virtualized containers lightweight and secure enough to be useful for workloads like serverless and agent sandboxes.\r\n\r\nStarting with benchmarks, we analyze the hidden costs and security implications of CoCo. We propose that, for single-purpose workloads, specialization is an essential next step toward the infrastructure \"pick-three\": speed, scale, and strong security. We dive into how tailored stacks - lightweight VMMs and guest OSes (via unikernels or customized Linux) - fit Confidential Computing context to restore VM-grade isolation with minimal overhead and attack surface. Finally, we highlight the challenges of managing and monitoring these specialized runtimes in production.", "startsAt": "2026-09-09T11:45:00", "endsAt": "2026-09-09T12:15:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "0b25d981-368c-4648-8ab6-e480d86991b4" ], "categoryItems": [ 447053, 446826, 447044 ], "questionAnswers": [ { "questionId": 134032, "answerValue": "https://sessionize.com/download/isgavos~J6rXBd28tT8FKi77Q8rVqS.pdf~lightweight_coco_kailun-qin.pdf" } ], "roomId": 80137, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1255192", "title": "Lunch (Not Provided) 🍜", "description": null, "startsAt": "2026-09-09T12:15:00", "endsAt": "2026-09-09T13:45:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [], "categoryItems": [ 485106 ], "questionAnswers": [], "roomId": 88856, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": false }, { "id": "1294797", "title": "Maintainer Meet-Up", "description": "The Maintainer Meetup is for CNCF Maintainers to share best practices, dive into contributing processes, and solve common problems across projects.", "startsAt": "2026-09-09T13:00:00", "endsAt": "2026-09-09T15:00:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [], "categoryItems": [ 485109 ], "questionAnswers": [], "roomId": 87761, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1311159", "title": "Project Pavilion | Wednesday PM Project Tables", "description": "Wednesday PM Project Tables | 13:15 - 15:30\r\n\r\nT-1\tOpenEverest\r\nT-2\tVolcano\r\nT-3 Cilium\r\nT-4\tDragonfly\r\nT-5\tCNCF Table\r\nT-6\tCommunity-Led Events\r\nT-7\tllm-d\r\nT-8\tspiderpool\r\nT-9\tOpenKruise\r\nT-10 hwameistor\r\nT-11 OpenStack (OpenInfra project)\r\nT-12 Safetensors (PyTorch project)\r\nT-13 kubean\r\n\r\n\r\nView the full project table directory here: https://www.lfopensource.cn/kubecon-cloudnativecon-openinfra-summit-pytorch-conference-china/features-add-ons/project-engagement/#project-table-directory", "startsAt": "2026-09-09T13:15:00", "endsAt": "2026-09-09T15:30:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [], "categoryItems": [ 485109 ], "questionAnswers": [], "roomId": 85878, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1224524", "title": "Turning Fragmented GPU Clusters Into One Elastic Compute Pool", "description": "As AI infrastructure evolves, fragmented GPU clusters and bursty AI workloads make large-scale multi-cluster operations increasingly difficult. \r\nIn this talk, we share how we built an Elastic Compute Pool that federates fragmented GPU capacity across clusters into a unified global scheduling domain.\r\nWhile Kubernetes Dynamic Resource Allocation (DRA) and Kueue provide a strong foundation for AI scheduling, we found critical gaps in applying them to global-scale GPU pools—particularly around application-level topology awareness, failover-aware placement, and cross-cluster resource modeling.\r\nWe will dive into how we extended DRA to address these limitations and how we unified the orchestration of training and inference workloads across hundred-thousand-GPU-scale infrastructure.\r\nAttendees will learn the architectural tradeoffs, scheduler and resource management extensions, and operational lessons required to make Kubernetes federation practical for AI-native compute platforms.\r\n", "startsAt": "2026-09-09T13:45:00", "endsAt": "2026-09-09T14:15:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "cf242c3d-a444-4485-a248-0236709431d7" ], "categoryItems": [ 447047, 446826, 447045 ], "questionAnswers": [ { "questionId": 134032, "answerValue": "https://sessionize.com/download/iplagfel~KNgWE9Ba4DTmsabWTzMdB9.pdf~kubecon-shanghai-2026-slide-0909.pdf" } ], "roomId": 80134, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1218005", "title": "Beyond Static Pods: Dynamic GPU Sharing and Low-Latency Model Switching for LLM Inference on K8s", "description": "Kubernetes revolutionized microservice orchestration, but Model-as-a-Service exposes a fundamental gap, where MaaS must handle thousands of heterogeneous models and extreme request variability on expensive GPU clusters while meeting strict TTFT/TPOT SLOs. At a large scale inference platform, traditional Kubernetes abstractions prove too coarse-grained: Pod-level model switching takes tens of seconds and idle GPUs remain occupied by inactive models.\r\n\r\nWe present Fast Model Actuation (FMA), an llm-d incubation project decoupling resource declaration from execution via Dual Pods. A server-requesting Pod holds GPU quota while a server-providing Pod runs inference, dynamically accessing the allocated GPU. Combined with vLLM sleep/wake for fast wake-up and a persistent Multi-Instance Launcher that creates inference subprocesses dynamically, switching drops from minutes to seconds, enabling time-division GPU sharing.\r\n\r\nWe cover the design, architecture, technical implementation and a demo.", "startsAt": "2026-09-09T13:45:00", "endsAt": "2026-09-09T14:15:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "782a0eb2-41fe-4175-9452-1ed725201aab", "4627a7ee-a0d9-4001-acfd-8cdabc022077" ], "categoryItems": [ 447047, 446827, 447045 ], "questionAnswers": [], "roomId": 80135, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1223545", "title": "To Cache or Not to Cache? A Tiered KVCache Storage System for Agent Scenarios", "description": "Abstract:\r\nKVCache reuse is a critical lever for reducing inference costs, with scenarios like OpenClaw and Code Agents achieving reuse rates exceeding 90%. While existing KVCache storage systems prioritize transmission throughput, they often overlook lifecycle management—specifically eviction and tiering policies—leaving the storage potential underutilized.\r\n\r\nIn this talk, we will dive deep into Unified Cache Manager's (UCM) lifecycle management framework, highlighting how heuristic-based dynamic access and eviction policies optimize cache utilization and further drive down inference costs.\r\n\r\nThe presentation contens:\r\n1. Workload Analysis: Data characteristics and retention patterns across OpenClaw, Code Agents, and general chat sessions.\r\n2. Predictive Modeling: Modeling capacity requirements and retention windows across diverse scenarios, models, and hardware architectures.\r\n3. Policy Implementation: heuristic-driven retention polices and performance gains.", "startsAt": "2026-09-09T13:45:00", "endsAt": "2026-09-09T14:15:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "f1bebe17-344f-4716-9b12-901b184b967f", "6b9fccdc-3e14-44f5-9d72-34cd95db79b9", "aac99729-6379-4968-ac02-a13cc6ea9dad" ], "categoryItems": [ 447046, 446824, 447045 ], "questionAnswers": [], "roomId": 80136, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1223326", "title": "From Disk Images to Container Images: Deploying bootc to Bare Metal with Kubernetes and Ironic", "description": "Bare metal provisioning traditionally relies on building and distributing qcow2 disk images. This results in build pipeline complexity, and limits hardware portability. With bootc (image-mode Linux), a container image contains a complete bootable operating system, including kernel, bootloader, etc. It is stored and distributed identically to any application container image, meaning image registries are available for distribution.\r\n\r\nThis session presents new OpenStack Ironic capabilities that allow Metal3 to provision bare metal servers directly from bootc container images, with no disk image built or distributed at any point.\r\n\r\nThe talk begins with a grounding in how container images are stored and how this relates to bootc images and OCI artefacts. Three architectural approaches to bare metal image delivery via Metal3 and Ironic are described. This results in a demonstration showing the end-to-end bootc provisioning flow, from Kubernetes custom resource creation to a deployed host.\r\n", "startsAt": "2026-09-09T13:45:00", "endsAt": "2026-09-09T14:15:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "b67a8b89-5f45-4650-9456-658b42f8800e" ], "categoryItems": [ 447053, 446826, 447044 ], "questionAnswers": [ { "questionId": 134032, "answerValue": "https://sessionize.com/download/cigyatne~hp6btQPKPph7mhbUMHzhnP.pdf~pres.pdf" } ], "roomId": 80137, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1293054", "title": "Helion's New CuTe DSL and Pallas Backends and How we Built Them via Agents", "description": "Helion is a Python DSL that raises kernel authoring above languages like CUDA and Triton: you describe a kernel at a high level, and a compiler and autotuner generate the device code. This talk introduces two new compiler backends.\r\n\r\nThe motivation is hardware heterogeneity and performance. Accelerators are diverging, and reaching state-of-the-art performance on each one normally means hand-writing kernels in that vendor's low-level language. The new backends lower a Helion kernel through CuteDSL for recent NVIDIA GPUs and through Pallas for TPUs. The same source targets different hardware without giving up that performance or rewriting per vendor.\r\n\r\nThe second half covers code generation by LLM agents. Because Helion sits above the languages it compiles to, a kernel takes fewer tokens to express and offers less surface area for mistakes, and the autotuner handles the performance tuning that agents do poorly. We cover where this holds up and where it doesn't.", "startsAt": "2026-09-09T13:45:00", "endsAt": "2026-09-09T14:15:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "02e37759-6364-4986-bac2-f67c808f0033", "6ac086c7-2c70-4b54-96ad-2e816c314f31" ], "categoryItems": [ 447047 ], "questionAnswers": [], "roomId": 85688, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": false, "isConfirmed": false }, { "id": "1222557", "title": "veRL: Extreme Optimization Practices for Ultra-Large-Scale MoE Models", "description": "Several leading internet companies (including ByteDance, JD.com, etc.) have built core RL training infrastructure based on veRL to support AI businesses. However, in practical deployment, rapid integration of new models, long-sequence training performance, and training stability remain key challenges.\r\n\r\nThis talk will introduce the latest technology and engineering practices of the veRL framework. You will learn:\r\n\r\n1. We have introduced expert parallelism and sequence parallelism partitioning strategies on top of PyTorch's native FSDP solution.\r\n2. Systematic optimizations for long-sequence training performance:\r\n- Asynchronous RL\r\n- Speculative inference acceleration\r\n- Extreme performance training with low precision\r\n3. Training stability:\r\nTraining collapse has long plagued the industry. After sustained and systematic efforts, we have finally achieved full alignment between training and inference.", "startsAt": "2026-09-09T14:30:00", "endsAt": "2026-09-09T15:00:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "45d59999-bcf1-4615-89cf-0be16523cd5a" ], "categoryItems": [ 447047, 446827, 447045 ], "questionAnswers": [], "roomId": 80134, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1223860", "title": "End-to-End Observability for LLM Inference: From Token to GPU", "description": "LLM inference is becoming the most expensive—and hardest to observe—workload on cloud-native clusters. Unlike stateless microservices, responses are streamed over tens of seconds; performance depends on GPU utilization, KV cache, and continuous batching; and token-level metrics like TTFT, TPOT, and ITL break the classic RED/USE model. A typical Prometheus + Trace stack only tells you the request \"succeeded\"—not why it was slow, or which layer was slow.\r\n\r\nThis talk shares our production practice of end-to-end observability for LLM inference, spanning the ingress, Inference Gateway, inference engine (vLLM/SGLang), model runtime, and GPU hardware. We show how to unify metrics across layers, correlate traces through streaming requests, and standardize on the OpenTelemetry GenAI semantic conventions. Three real incidents—a P99 spike, a \"busy-but-idle\" GPU, and a KV cache OOM—demonstrate how this cuts time-to-resolve from hours to minutes, with a reusable open-source toolchain and configs.", "startsAt": "2026-09-09T14:30:00", "endsAt": "2026-09-09T15:00:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "4dbc9a9e-ef1f-4e4d-b0d1-0c8442cf1b71", "7b380a7c-a033-417e-af64-6741a72e400e", "50e09a13-2711-4c71-b987-611717466667" ], "categoryItems": [ 447050, 446826, 447045 ], "questionAnswers": [ { "questionId": 134032, "answerValue": "https://sessionize.com/download/ifiymay~Dur7z1Dbqz7DLEgf5TnNjJ.pdf~end-to-end-observability-for-llm-inference-from-token-to-gpu-en.pdf" } ], "roomId": 80135, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1225189", "title": "Redesigning OpenStack i18n for the AI Era: Weblate & Zero-GPU AI", "description": "\"No open source tool lasts forever.\" With Zanata reaching end-of-life, OpenStack i18n SIG migrated decades of translation artifacts across 50+ languages to Weblate. Rather than a tool replacement, we saw this as a challenge to redefine the future of open-source internationalization in the AI era.\r\n\r\nThis session shares our journey of transforming a legacy translation system into a resilient, automated, and AI-ready pipeline at scale. \r\n\r\n(1) Migration Strategies: Preserving data integrity across 50+ languages during large-scale workflow transitions\r\n\r\n(2) Resilient Architecture: Zuul-powered translation sync and robust failure recovery\r\n\r\n(3) Zero-GPU AI Integration: Running open-source LLMs on cloud instances (up to 16 cores) for commit-time draft translations—zero GPU dependency.\r\n\r\nThrough the OpenInfra University Partnership Program, we mentored Korean students to onboard as active contributors, demonstrating how i18n modernization efforts directly drive sustainable ecosystem growth.", "startsAt": "2026-09-09T14:30:00", "endsAt": "2026-09-09T15:00:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "e051c049-14f7-480f-8edb-c02677855aa3", "a27033da-3f4a-49ae-b6d5-8a686df7e39e" ], "categoryItems": [ 447055, 446825, 447044 ], "questionAnswers": [ { "questionId": 134032, "answerValue": "https://sessionize.com/download/icwavver~ayprGQRRaKPe4WD7h6RkmE.pdf~redesigning-openstack-i18n-for-the-ai-era_-weblate-zero-gpu-ai.pdf" } ], "roomId": 80136, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1220661", "title": "From Cloud to AI: How Kata Containers 4.0 Reinvents the Sandbox for the Agent Era", "description": "AI agents need more than Linux. Kata Containers 4.0 delivers a universal secure sandbox for the agent era. A new Rust-based default runtime replaces Go for superior safety and speed. Multi-hypervisor support — QEMU, Dragonball, Cloud Hypervisor, Firecracker — adapts to any infrastructure. Template-based instant startup achieves sub-100ms sandbox launch, enabling elastic agent scaling at cloud speed. Confidential Computing via Intel TDX, AMD SEV-SNP, Hygon CSV, and DCU protects sensitive workloads at hardware level. And for the first time, Kata breaks beyond Linux: Android, OSWorld, and Windows sandboxes unify under one runtime, powering diverse AI agent environments securely and instantly. The future of sandboxing is multi-world — and it starts with Kata 4.0.", "startsAt": "2026-09-09T14:30:00", "endsAt": "2026-09-09T15:00:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "87a1caef-272a-4dd5-9390-cd7e06480345", "82737250-2ba7-402d-87b4-78bb8d7ac2ff" ], "categoryItems": [ 447053, 446826, 447044 ], "questionAnswers": [], "roomId": 80137, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1222012", "title": "Multi-cluster Orchestration System: Karmada Updates and Use Cases", "description": "Karmada (Kubernetes Armada) is a Kubernetes management system that enables you to run your cloud-native applications across multiple Kubernetes clusters and clouds. \r\n\r\nIn this presentation, the maintainer of the Karmada project will share: \r\n\r\n- A Brief introduction to Karmada.\r\n- New features over the last year\r\n Application Priority Scheduling\r\n Stateful Application Cluster Failover\r\n Federated ResourceQuota Enforcement\r\n Workload Affinity\r\n AI Jobs Scheduling Enhancements\r\n- Real-world case studies\r\n- Overview of the community \r\n- Roadmap\r\n- QA\r\n", "startsAt": "2026-09-09T14:30:00", "endsAt": "2026-09-09T15:00:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "1afdc8f0-ebbb-4f55-be52-64bf0815b421", "ae4766ae-8a22-4193-bb37-a9bc23fa67f0" ], "categoryItems": [ 485108, 447045, 486116 ], "questionAnswers": [], "roomId": 85688, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1255191", "title": "Coffee Break ☕", "description": null, "startsAt": "2026-09-09T15:00:00", "endsAt": "2026-09-09T15:30:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [], "categoryItems": [ 485106 ], "questionAnswers": [], "roomId": 85878, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": false }, { "id": "1222763", "title": "vLLM KV Cache Management: From Cache Reuse to Agent Scenario Optimization", "description": "As LLM Agents proliferate, the conflict between surging KV Cache and limited VRAM in multi-turn dialogues and tool-chaining has become a key bottleneck. This session shares proven vLLM practices to support longer contexts and higher concurrency under resource constraints.\r\nTo address this, vLLM leverages two core mechanisms: Prefix Caching for block-hash-based reuse of common task data (prompts, RAG, history), and KV Offloading to maximize utilization via a three-tier fallback system (LRU, CPU offloading, and ARC policies).\r\nWe also explore future roadmap: semantic-aware KV compression based on attention scores, priority-based scheduling for Agent task dependencies, and cross-session prefix sharing. Attendees will understand vLLM’s KV Cache trade-offs, master metrics like Prefix Cache Hit Rate, and preview the evolution of cache optimization for Agent-driven workloads.", "startsAt": "2026-09-09T15:30:00", "endsAt": "2026-09-09T16:00:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "edb71874-107d-4c2b-a50f-72fe29a26d23", "6b5d35dd-d75f-4e76-908a-bc21bd7d772e" ], "categoryItems": [ 447047, 446827, 447045 ], "questionAnswers": [], "roomId": 80134, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": false }, { "id": "1219307", "title": "Non-Invasive AI Agent Observability With OBI", "description": "AI applications have evolved from simple LLM calls to sophisticated agent workflows involving tool invocations, MCP interactions, vector embeddings, and RAG. A single agent task may trigger dozens of LLM calls and cross-service tool executions, making debugging extremely challenging.\r\nThis talk introduces how OpenTelemetry eBPF Instrumentation (OBI) extends observability capabilities for AI agents. By intercepting traffic at the network layer via eBPF, OBI transparently captures agent behavior — including MCP JSON-RPC tool calls, embedding and reranking operations, and multi-turn LLM-tool interactions. We will demonstrate how eBPF detects MCP Streamable HTTP, extracts tool calls from LLM responses (OpenAI, Anthropic, Gemini, and Qwen), and how the resulting telemetry data aligns with OpenTelemetry GenAI semantic conventions. Attendees will learn how zero-instrumentation observability enables comprehensive and transparent monitoring of complex AI agent systems running on Kubernetes.", "startsAt": "2026-09-09T15:30:00", "endsAt": "2026-09-09T16:00:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "36602fff-5c9a-4b8a-bbfd-16168f1c9a0a", "a3c542ee-6ee8-44e3-9588-f2f4c1faa06d" ], "categoryItems": [ 447050, 446827, 447044 ], "questionAnswers": [ { "questionId": 134032, "answerValue": "https://sessionize.com/download/islawkes~pgcj26mJd8aau8pRoySAGc.pdf~non-invasive-ai-agent-observability-with-obi-kubecon-extended.pdf" } ], "roomId": 80135, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1225138", "title": "The State of the PyTorch Ecosystem in 2026: Global Trends and China’s Rising Role", "description": "In 2026, PyTorch stands at the center of a rapidly expanding global AI ecosystem. Through a data-driven analysis of activity across GitHub, Hugging Face, and related open platforms, this talk reveals a clear upward trajectory in contributions from China—spanning code, models, tools, and community engagement. Yet as domestic efforts in AI frameworks and hardware accelerate, the risk of fragmentation grows. We argue that China’s homegrown AI infrastructure should actively converge with mainstream ecosystems like PyTorch, not to replace them, but to co-develop interoperable standards that benefit the entire global community. The path forward lies not in isolated stacks, but in shared foundations.", "startsAt": "2026-09-09T15:30:00", "endsAt": "2026-09-09T16:00:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "2509fe84-b256-432f-accd-e786bef7734c" ], "categoryItems": [ 447046, 446827, 447045 ], "questionAnswers": [], "roomId": 80136, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1222331", "title": "Write Once, Run Anywhere: PyTorch’s Generalization Journey", "description": "As AI hardware continues to diversify, PyTorch faces a growing challenge: many of its APIs, runtime interface, and test infrastructures remain fragmented and backend-specific, making it difficult for users to write portable code and for developers to support new hardware consistently.\r\nThis talk presents our ongoing generalization effort to make PyTorch truly “write once, run anywhere.” We cover three key areas: unifying module-level APIs such as Autocast, Inductor, and Graph to lower the cost of supporting new hardware; introducing torch.accelerator, a unified runtime API for device, stream, and memory management; and restructuring test infrastructure to validate correctness across multiple backends.\r\nTogether, these efforts reduce per-backend maintenance burden, improve test coverage uniformity, and establish a scalable foundation for integrating new hardware into PyTorch with significantly less effort and greater confidence.\r\n", "startsAt": "2026-09-09T15:30:00", "endsAt": "2026-09-09T16:00:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "c3488095-867e-43ca-85d8-704cb71ace2f", "fe44db8a-c695-465f-b248-d58aba846c72" ], "categoryItems": [ 447047, 446827, 447045 ], "questionAnswers": [ { "questionId": 134032, "answerValue": "https://sessionize.com/download/iksavpei~hizufp2XJop4TReHNPCj8d.pdf~pytorch-conference-china-2026-write-once-run-anywhere-pytorchs-generalization-journey.pdf" } ], "roomId": 80137, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1225125", "title": "Solving Industrial Challenges with KubeEdge: A Post-Graduation Report", "description": "Following its graduation within the CNCF, KubeEdge has solidified its position as the premier platform for extending Kubernetes to the edge. In this session, project maintainers will explore KubeEdge's evolution, offering a deep dive into the core architecture that enables efficient management of edge workloads.\r\nAttendees will gain insights from real-world deployments across diverse sectors, including Smart Cities, Industrial IoT (IIoT), Edge AI, Robotics, and Retail. Beyond success stories, the talk will cover critical technical updates, including the newly introduced Certified KubeEdge conformance test, recent technological advancements, and the latest updates on community governance.\r\n", "startsAt": "2026-09-09T15:30:00", "endsAt": "2026-09-09T16:00:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "23394802-299a-42d2-9aa2-4eb5afe2894d", "ff2aa64c-9b1b-48fc-b5d9-a85e5d8694c2", "f326d9b7-e8df-4757-8923-58bf9bb6137c" ], "categoryItems": [ 485108, 447045, 486111 ], "questionAnswers": [], "roomId": 85688, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1214156", "title": "vLLM-Helion: SOTA LLM Performance by advanced autotuning and fine-grained dispatching", "description": "High-performance LLM inference demands kernels that are not only algorithmically optimal but also precisely tuned to the hardware and workload at hand. We present vLLM's integration with Helion, PyTorch's Python-embedded kernel authoring DSL.\r\n\r\nvLLM-Helion is built around two key ideas: (1) an advanced offline autotuning infrastructure that searches the configuration space across a comprehensive set of input shapes representative of real production workloads, and ships the results as pre-tuned platform-specific configs, eliminating runtime tuning overhead entirely; and (2) a fine-grained runtime dispatching mechanism that selects the optimal pre-tuned configuration based on actual input dimensions at each kernel invocation, rather than relying on a single \"one-size-fits-all\" config per platform.\r\n\r\nTogether, these techniques enable vLLM to achieve state-of-the-art kernel performance on many supported hardwares.\r\n", "startsAt": "2026-09-09T16:15:00", "endsAt": "2026-09-09T16:45:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "1ca6dcf8-473e-4737-a075-baabb268dab6" ], "categoryItems": [ 447047, 446827, 447044 ], "questionAnswers": [ { "questionId": 134032, "answerValue": "https://sessionize.com/download/impaldey~Lu9nPHGesXnMGADsrTkByA.pdf~vllm-helion-pytorch-conference-china-2026.pdf" }, { "questionId": 134033, "answerValue": "https://github.com/redhat-et/vllm-helion" } ], "roomId": 80134, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1220647", "title": "Why Your TTFT Lies: Diagnosing PD-Disaggregated LLM Inference with Minimal Cross-Layer Metrics", "description": "Why is inference slow? You have metrics from frameworks/engines/GPUs, but they don't tell you what's wrong. Prefill queueing? Decode memory‑bound? KV transfer latency? GPU/network bottleneck?\r\nWith PD‑disaggregated architecture, the problem worsens: different stages behave completely differently, yet observability can't distinguish them.\r\nRoot cause: Production LLM inference is a multi‑stage system (Gateway, engines, GPUs, network) — but metrics are siloed. PD‑disaggregation amplifies this: each stage has its own performance behavior, yet monitoring can't tell them apart.\r\n\r\nWe present a unified observability method for PD‑disaggregated systems. Using a few high-impact metrics, we build a direct mapping: metrics → root cause → tuning action:\r\n- PD‑aware view linking gateway, engine internals (KV Cache, batching, scheduler), infra, GPU & network metrics.\r\n- upgrade from *dashboard watching* to interpretable diagnosis → actionable tuning (P:D ratio, batch size, KV policy, RDMA/topology).", "startsAt": "2026-09-09T16:15:00", "endsAt": "2026-09-09T16:45:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "50e09a13-2711-4c71-b987-611717466667", "26415447-f9a1-4909-a6df-6695199e792f" ], "categoryItems": [ 447050, 446826, 447045 ], "questionAnswers": [ { "questionId": 134032, "answerValue": "https://sessionize.com/download/idyatweu~QMcLoJtAyViBA2cZx3Mwny.pptx~kubecon-china-2026-why-your-ttft-lies.pptx" } ], "roomId": 80135, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1218010", "title": "What We Learned Securing AI Agents at Scale on Multi-Tenant Kubernetes Clusters", "description": "Giving every engineer a personal AI agent sounds great until you put hundreds of them on a shared Kubernetes cluster. Each agent executes arbitrary code, calls external tools via MCP, holds cloud credentials, and makes network decisions on its own. Standard Kubernetes multi-tenancy (namespace boundaries, RBAC, NetworkPolicy) was designed for cooperative workloads, not for autonomous agents that can act in unpredictable ways.\r\n\r\nWe built a multi-tenant agent platform serving hundreds of engineers for daily development and oncall incident response. This talk covers the defense-in-depth architecture we arrived at: Kata Containers giving each agent a VM with its own kernel, Agent ID assigning each agent instance a unique cloud identity via OIDC federation so it can only access data scoped to its owner, NetworkPolicy blocking IMDS and lateral movement, ValidatingAdmissionPolicy preventing exec breakouts, and governance guardrails that enforce blast-radius limits with full audit logging.", "startsAt": "2026-09-09T16:15:00", "endsAt": "2026-09-09T16:45:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "38662798-6e4b-4de9-96b6-639fb6021de1" ], "categoryItems": [ 447046, 446827, 447045 ], "questionAnswers": [ { "questionId": 134032, "answerValue": "https://sessionize.com/download/ilzawiv~4MTg7dwnN4m6TLkGTZpAu7.pdf~what-we-learned-securing-ai-agents-at-scale-on-multi-tenant-kubernetes-clusters.pdf" } ], "roomId": 80136, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1223308", "title": "Kubernetes DRA Architecture: Scheduling, Status, and Topology at Scale", "description": "AI/HPC workloads need more than “two GPUs”: they need GPUs, NICs, NUMA locality, and PCIe/fabric relationships. This session presents Kubernetes Dynamic Resource Allocation (DRA) in v1.36 as one architecture across entry points, resource modeling, scheduler allocation, binding, kubelet preparation, status, health, and access boundaries. One diagram shows how the main DRA KEPs(**14 DRA-related KEPs** were updated in v1.36 release cycle) fit together.\r\n\r\nThe focus is topology-aware scheduling for supernode systems, where NUMA, PCIe, NVLink, or fabric placement determines performance. We will also present an IMEX daemon/channel assignment case study: the design evolved from distributed status updates with conflicts, to centralized updates, and finally to distributed updates on a topology-aligned state object, reducing assignment latency from minutes to seconds at thousand-device scale.", "startsAt": "2026-09-09T16:15:00", "endsAt": "2026-09-09T16:45:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "dd481bd7-e625-4e56-a76b-4baa54d457a5", "27e5f808-79cf-4a54-9512-d572400f9e67" ], "categoryItems": [ 447054, 446826, 447045 ], "questionAnswers": [ { "questionId": 134032, "answerValue": "https://sessionize.com/download/kqaxyec~47fBae5CA846z2Fj7Xfbwx.pdf~kubernetes-dra-architecture-v1.1.pdf" }, { "questionId": 134033, "answerValue": "https://github.com/NVIDIA/k8s-dra-driver-gpu" } ], "roomId": 80137, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true }, { "id": "1222060", "title": "Volcano: A Unified Scheduling Platform for Cloud Native AI", "description": "AI infrastructure is moving beyond a single workload model. Training, inference, and agent workloads often run in separate systems today, making resource sharing, queue management, topology placement, and accelerator abstraction harder to operate.\r\n\r\nVolcano is evolving as a unified scheduling platform for the AI lifecycle. Volcano-Global extends training jobs across clusters. Kthena is becoming a production-ready LLM serving layer, with rolling updates, router observability, prefix-cache routing, vLLM/SGLang support, and upcoming autoscaling, session affinity, and token-cost-aware balancing. AgentCube brings agent runtimes and code interpreters into Kubernetes with microVM session routing, warm pools, PicoD execution, and SDKs.\r\n\r\nAt the infrastructure layer, Volcano is improving gang preemption, sharding, namespace queues, throughput, GPU sharing, and HyperNode topology. This session shares the maintainer roadmap and how these pieces converge into one cloud native AI scheduling stack.", "startsAt": "2026-09-09T16:15:00", "endsAt": "2026-09-09T16:45:00", "isServiceSession": false, "isPlenumSession": false, "speakers": [ "4e85174f-e168-489c-8626-d2c762b3fc99", "a915de2e-ee34-4e5a-a8ef-2a50bcf70e33" ], "categoryItems": [ 485108, 447045, 486117 ], "questionAnswers": [], "roomId": 85688, "liveUrl": null, "recordingUrl": null, "status": "Accepted", "isInformed": true, "isConfirmed": true } ], "speakers": [ { "id": "dd6c32e8-7b29-4f89-9026-344f951594b8", "firstName": "Amir", "lastName": "Afzali", "bio": "Software engineer working on distributed machine learning infra", "tagLine": "Software Engineer, Meta PyTorch", "profilePicture": "https://cdn.sessionize.com/image/0e56-400o400o1-HoobYns1JwWYJuzsbY3LdQ.png", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/amirafzali", "linkType": "LinkedIn" } ], "sessions": [ 1223450 ], "fullName": "Amir Afzali", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Meta" }, { "questionId": 123918, "answerValue": "Software Engineer" } ] }, { "id": "e355bcd9-b499-4179-8a69-5d6a850b4d2e", "firstName": "Baofa", "lastName": "Fan", "bio": "Baofa Fan (GitHub @carlory) is an active reviewer of the Kubernetes, Kubernetes-sigs and Kubernetes-csi organization, currently mainly on sig-storage. \r\n\r\nAnd He is also a reviewer of the Karmada project which focus on the multi-cluster area.", "tagLine": "DaoCloud, IT Engineer ", "profilePicture": "https://cdn.sessionize.com/image/44e5-400o400o1-5G9MZqkZNcA6UmA5H9nTn.jpeg", "isTopSpeaker": false, "links": [ { "title": "Company Website", "url": "https://www.daocloud.io", "linkType": "Company_Website" }, { "title": "GitHub", "url": "https://github.com/carlory", "linkType": "Other" }, { "title": "Sessionize", "url": "https://sessionize.com/fan-baofa", "linkType": "Sessionize" } ], "sessions": [ 1193509 ], "fullName": "Baofa Fan", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "DaoCloud" }, { "questionId": 123918, "answerValue": "IT Engineer " } ] }, { "id": "516cce70-ba6b-406e-845c-4e0ea6471583", "firstName": "Bertrand", "lastName": "Souville", "bio": "Bertrand Souville is Senior Architect at DOCOMO Euro-Labs and involved with OpenStack since the Liberty release. He supports the virtualization platforms for both RAN and Core Network, bridging the gap between development requirements and global standards.", "tagLine": "Senior Architect", "profilePicture": "https://cdn.sessionize.com/image/f19b-400o400o1-P9KFmWhUdWcD6HgkCTep17.png", "isTopSpeaker": false, "links": [], "sessions": [ 1224932 ], "fullName": "Bertrand Souville", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "DOCOMO Euro-Labs" }, { "questionId": 123918, "answerValue": "Senior Architect" } ] }, { "id": "3872d9f6-ea0a-4421-a994-a803e688411f", "firstName": "Bin", "lastName": "Yang", "bio": "Bin leads Intel's LLM inference optimization work, responsible for delivering high-performance, hardware-accelerated inference solutions on Intel platforms, with a focus on optimizing and upstreaming these capabilities\r\n into various open-source communities and the broader inference ecosystem to lower the cost of AI serving across Intel hardwares.", "tagLine": "LLM KV Cache expert making AI inference faster and more cost-efficient on Intel platforms", "profilePicture": "https://cdn.sessionize.com/image/663d-400o400o1-RbMZvRenjr9322V3gDSgLj.png", "isTopSpeaker": false, "links": [ { "title": "Sessionize", "url": "https://sessionize.com/bin-yang", "linkType": "Sessionize" } ], "sessions": [ 1303982 ], "fullName": "Bin Yang", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Intel" }, { "questionId": 123918, "answerValue": "Senior Engineer" } ] }, { "id": "f5200c5c-d81a-488c-b57a-520e94c626be", "firstName": "Bingshen", "lastName": "Wang", "bio": "Bingshen Wang is a Senior Engineer in Alibaba Could, a maintainer of KubeSkoop/Terway/OpenYurt, and a contributor of Kubernetes/Containerd. He mainly focuses on container networking and runtime, and has many years of experience around managing Alibaba Cloud Kubernetes clusters.\r\nHe used to be a speaker in KubeCon conferences and some other meetup or virtual submit such as GOTC and sig/cloud-provider-alibaba.", "tagLine": "AlibabaCloud, Senior Engineer", "profilePicture": "https://cdn.sessionize.com/image/544f-400o400o1-SE5iBpWPgiFSnK2h6LoHU4.png", "isTopSpeaker": false, "links": [ { "title": "Sessionize", "url": "https://sessionize.com/bingshen-wang", "linkType": "Sessionize" } ], "sessions": [ 1221856 ], "fullName": "Bingshen Wang", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Alibaba Cloud" }, { "questionId": 123918, "answerValue": "Software Engineer" } ] }, { "id": "150879da-174d-4ebf-aa1c-ebebdbd5e20d", "firstName": "Bo", "lastName": "Chu", "bio": "Bo Chu works at Alibaba Cloud, focusing on cloud-native AI infrastructure. He is deeply engaged in the AI ecosystem — from workload orchestration on Kubernetes, to distributed model training, to large-scale inference serving. An active contributor to the open source community, he closely follows advancements in cloud-native and AI technologies.", "tagLine": "Senior Engineer, Alibaba Cloud", "profilePicture": "https://cdn.sessionize.com/image/f93f-400o400o1-BSYGpJngQD6sYsBvkqWoz4.png", "isTopSpeaker": false, "links": [ { "title": "Sessionize", "url": "https://sessionize.com/bo-chu", "linkType": "Sessionize" } ], "sessions": [ 1224503 ], "fullName": "Bo Chu", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Alibaba Cloud" }, { "questionId": 123918, "answerValue": "Senior Software Engineer" } ] }, { "id": "93496157-eb91-4ced-8ff2-51d506e10493", "firstName": "Bo", "lastName": "Fu", "bio": "Bo Fu is a Senior Product Manager for Google Kubernetes Engine.", "tagLine": "Google, Senior Product Manager", "profilePicture": "https://cdn.sessionize.com/image/6f2f-400o400o1-4Wvqt1GwHs1ZQXFYTCnULo.jpg", "isTopSpeaker": false, "links": [], "sessions": [ 1210983 ], "fullName": "Bo Fu", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Google" }, { "questionId": 123918, "answerValue": "Senior Product Manager" } ] }, { "id": "ea195666-322a-48a9-9402-c6a944a4c4b1", "firstName": "Bo Kang", "lastName": "Li", "bio": "BoKang Li is a senior engineer on Container Service for Kubernetes team of Alibaba Cloud. He has a primary focus on container networks and connectivity solutions, and has gained extensive experience in use container networking.", "tagLine": "Senior Engineer, Alibaba Cloud", "profilePicture": "https://cdn.sessionize.com/image/d7c4-400o400o1-SqVMDVBmgsrMXLiYv7mpSu.jpg", "isTopSpeaker": false, "links": [ { "title": "Sessionize", "url": "https://sessionize.com/bo-kang-li", "linkType": "Sessionize" } ], "sessions": [ 1221856 ], "fullName": "Bo Kang Li", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Alibaba Cloud" }, { "questionId": 123918, "answerValue": "Senior Engineer" } ] }, { "id": "45d59999-bcf1-4615-89cf-0be16523cd5a", "firstName": "Carson", "lastName": "Wu", "bio": "I am a reinforcement learning systems engineer at Huawei Ascend Computing, dedicated to advancing efficient RL training on AI accelerators. Since April 2024, I have been building and optimizing large-scale RL training infrastructure, with a focus on PyTorch-based distributed frameworks. I am an Committer for veRL and MindSpeed-RL, two open-source projects that enable efficient LLM post-training and RL scaling.", "tagLine": "Huawei Technologies Co., Ltd.", "profilePicture": "https://cdn.sessionize.com/image/6ab7-400o400o1-nt7R5MBDqTgehnrsnrgokx.jpg", "isTopSpeaker": false, "links": [], "sessions": [ 1222557 ], "fullName": "Carson Wu", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Huawei Technologies Co., Ltd." }, { "questionId": 123918, "answerValue": "reinforcement learning systems engineer " } ] }, { "id": "4e85174f-e168-489c-8626-d2c762b3fc99", "firstName": "Chen", "lastName": "Zicong", "bio": "Zicong Chen is the CNCF project Volcano maintainer and a Software Engineer at Huawei Cloud specializing in cloud-native scheduling. He possesses deep expertise in this domain and is dedicated to solving the scheduling challenges posed by complex workloads like AI/HPC. As an active community leader, he regularly hosts the Volcano community meetings and has spoken at technical conferences including KCD Beijing, GOSIM Hangzhou, Kubecon EU 26 etc. ", "tagLine": "CNCF Volcano Maintainer, LeaderWorkerSet Contributor & R&D Engineer at Huawei Cloud", "profilePicture": "https://cdn.sessionize.com/image/fc90-400o400o1-Ep2G4FoR13ChNbAwU38HHp.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/jesse-stutler/", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://www.huawei.com/en/", "linkType": "Company_Website" }, { "title": "Sessionize", "url": "https://sessionize.com/chen-zicong", "linkType": "Sessionize" } ], "sessions": [ 1222146, 1222060 ], "fullName": "Chen Zicong", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Huawei Technologies Co., Ltd." }, { "questionId": 123918, "answerValue": "Volcano Maintainer, Software Engineer at Huawei Cloud" } ] }, { "id": "f4827ef5-9d40-44e9-8f70-be0c21c27af2", "firstName": "Chenyu", "lastName": "Zhang", "bio": "Chenyu Zhang is a software engineer at Broadcom, focusing on the development and maintenance of the CNCF projects Harbor and Dragonfly, with additional experience in DevOps and cloud-native technologies.", "tagLine": "Harbor Maintainer", "profilePicture": "https://cdn.sessionize.com/image/6812-400o400o1-wQX6qovMKJnLB7WTxGxMy8.png", "isTopSpeaker": false, "links": [], "sessions": [ 1174542 ], "fullName": "Chenyu Zhang", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "VMware by Broadcom" }, { "questionId": 123918, "answerValue": "Software Engineer" } ] }, { "id": "cf242c3d-a444-4485-a248-0236709431d7", "firstName": "ChongKang", "lastName": "Tan", "bio": "Chongkang Tan is a Staff Engineer at Ant Group, where his work focuses on efficient GPU, sandbox, and container compute services for agentic applications. His technical interests include container compute platforms, distributed cluster architecture, and heterogeneous computing. \r\nPreviously, he was a senior researcher at Lenovo, working across cloud platforms, edge computing, and operating systems. \r\nHe has spoken at KubeCon, QCon, BDTC, and other technical conferences.", "tagLine": "Staff Engineer,AntGroup", "profilePicture": "https://cdn.sessionize.com/image/e715-400o400o1-VRPm3YVrQvoCTXDGZLMenW.jpg", "isTopSpeaker": false, "links": [], "sessions": [ 1224524 ], "fullName": "ChongKang Tan", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Ant Group" }, { "questionId": 123918, "answerValue": "Staff Engineer" } ] }, { "id": "2a4df937-4b16-4b99-b27d-a07841c63fec", "firstName": "Chris", "lastName": "Aniszczyk", "bio": "Chris Aniszczyk is an open source executive and engineer with a passion for building a better world through open collaboration. He's currently a CTO at the Linux Foundation focused on developer relations and running the Open Container Initiative (OCI) / Cloud Native Computing Foundation (CNCF). Furthermore, he's a partner at Capital Factory where he focuses on mentoring, advising and investing in open source and infrastructure focused startups.\r\n\r\nAt Twitter, he created their open source program and led their open source efforts. For many years he served on the Eclipse Foundation's Board of Directors representing the committer community and the Java Community Process (JCP) Executive Committee. In a previous life, he bootstrapped a consulting company, made many mistakes, lead and hacked on many eclipse.org and Linux related projects", "tagLine": null, "profilePicture": "https://cdn.sessionize.com/image/cd5a-400o400o1-aS8Q5bY9F6o69zDPeD4KMg.jpg", "isTopSpeaker": true, "links": [], "sessions": [ 1312064 ], "fullName": "Chris Aniszczyk", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "The Linux Foundation" }, { "questionId": 123918, "answerValue": "CTO, Cloud and Infrastructure" }, { "questionId": 138874, "answerValue": "https://events.linuxfoundation.org/wp-content/uploads/2022/02/linux-foundation-vert-color.svg" } ] }, { "id": "a27033da-3f4a-49ae-b6d5-8a686df7e39e", "firstName": "DaGyeong", "lastName": "Kim", "bio": "DaGyeong Kim is interested in OpenStack and currently contributing to the OpenStack i18n team's Weblate migration project.", "tagLine": "Independent, Open Source Contributor", "profilePicture": "https://cdn.sessionize.com/image/1fd4-400o400o1-GzYAcJbmgUpYX1jEeL3hHN.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/dagyeong-kim-058776202/", "linkType": "LinkedIn" } ], "sessions": [ 1225189 ], "fullName": "DaGyeong Kim", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Independent" }, { "questionId": 123918, "answerValue": "Infrastructure Engineer" } ] }, { "id": "8f36d40e-c5d9-45fb-9982-88829e7eab04", "firstName": "Dandan", "lastName": "Liang", "bio": "Dandan Liang received her Ph.D. degree from the University of Southampton. After that, she was a Research Fellow at the University of Surrey. Then she joined Huawei Technologies as a Senior Engineer in wireless communications and Wi-Fi standardization. She is now an Associate Professor at Peng Cheng Laboratory. She has published 18 papers, filed 76 patents, and been recognised as a Kubeastronaut. Her research interests include cross-layer system design and cloud native network architecture.", "tagLine": "PCL, Associate Professor, Kubestronaut", "profilePicture": "https://cdn.sessionize.com/image/9ad0-400o400o1-w3jdnZdTmmXYdtntA7ycZU.png", "isTopSpeaker": false, "links": [ { "title": "Company Website", "url": "https://www.pcl.ac.cn/", "linkType": "Company_Website" }, { "title": "Sessionize", "url": "https://sessionize.com/DandanLIANG", "linkType": "Sessionize" } ], "sessions": [ 1224457 ], "fullName": "Dandan Liang", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Pengcheng Laboratory" }, { "questionId": 123918, "answerValue": "Associate Professor" } ] }, { "id": "4b82436a-947c-45bb-9d5d-3b94c02c8faa", "firstName": "Daniel", "lastName": "Oh", "bio": "Java Champion, CNCF Ambassador & TAG DevEX Co-Chair, Microsoft MVP, Developer Advocate, Technical Marketing, Keynote Speaker, Published Author", "tagLine": "Senior Principal Developer Advocate", "profilePicture": "https://cdn.sessionize.com/image/5b8c-400o400o1-WGaguRyvA6idCeTRfHC4my.png", "isTopSpeaker": false, "links": [ { "title": "X (Twitter)", "url": "https://twitter.com/danieloh30", "linkType": "Twitter" }, { "title": "LinkedIn", "url": "https://www.linkedin.com/in/daniel-oh-083818112/", "linkType": "LinkedIn" }, { "title": "Blog", "url": "https://www.youtube.com/@danieloh30", "linkType": "Blog" }, { "title": "Company Website", "url": "https://www.redhat.com", "linkType": "Company_Website" }, { "title": "Sessionize", "url": "https://sessionize.com/danieloh", "linkType": "Sessionize" } ], "sessions": [ 1180062 ], "fullName": "Daniel Oh", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Red Hat" }, { "questionId": 123918, "answerValue": "Senior Principal Developer Advocate" } ] }, { "id": "8b0bcad4-0594-4ae8-a240-891d6b34e55f", "firstName": "Di", "lastName": "Xu", "bio": "Currently, Di Xu serves as a Tech Lead at Xiaohongshu, focusing on building reliable and scalable infrastructure for cloud native, AI, and agent applications. He is a long-time Kubernetes contributor and cloud native community leader. He is also a CNCF Ambassador and has served as a track chair for multiple KubeCon + CloudNativeCon events.", "tagLine": "CNCF Ambassador | Kubernetes Member | Open Source Enthusiast", "profilePicture": "https://cdn.sessionize.com/image/5eeb-400o400o1-Xw1yF5xXX3zNVhioWePcWB.jpg", "isTopSpeaker": true, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/pub/di-xu/71/991/7a1", "linkType": "LinkedIn" } ], "sessions": [ 1309269 ], "fullName": "Di Xu", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Xiaohongshu" }, { "questionId": 123918, "answerValue": "CNCF Ambassador | KubeCon Co-Chair" }, { "questionId": 138874, "answerValue": "https://events.linuxfoundation.org/wp-content/uploads/2026/08/Xiaohongshu_2013.svg" } ] }, { "id": "a915de2e-ee34-4e5a-a8ef-2a50bcf70e33", "firstName": "DongYang", "lastName": "Wang", "bio": "Dongyang Wang is a Cloud-Native AI Infrastructure Engineer at SenseTime and a core contributor to the Volcano project, a CNCF-hosted system for high-performance scheduling on Kubernetes. He specializes in large-scale GPU and NPU scheduling, building intelligent systems that orchestrate tens of thousands of accelerators for distributed AI workloads.", "tagLine": "Cloud-Native AI Infrastructure Engineer at SenseTime, building intelligent scheduling systems for large-scale compute", "profilePicture": "https://cdn.sessionize.com/image/2619-400o400o1-M2fGmRABGn8bjCiuYhe3me.jpg", "isTopSpeaker": false, "links": [], "sessions": [ 1222060 ], "fullName": "DongYang Wang", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "SenseTime" }, { "questionId": 123918, "answerValue": "Cloud Native AI Infrastructure Engineer" } ] }, { "id": "02e37759-6364-4986-bac2-f67c808f0033", "firstName": "Dunfan", "lastName": "Lu", "bio": "I am a Software Engineer at Meta in the PyTorch compiler team. I am actively working on PyTorch and Helion. I have previously also contributed to taichi (https://github.com/taichi-dev/taichi) which is an open-source GPU programming language for computer graphics purposes. ", "tagLine": "Software Engineer - Meta", "profilePicture": "https://cdn.sessionize.com/image/2a8f-400o400o1-Y41cM9tar6MUAUiFQHTWFT.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/dunfan-lu-45b005156/", "linkType": "LinkedIn" }, { "title": "Sessionize", "url": "https://sessionize.com/dunfan-lu", "linkType": "Sessionize" } ], "sessions": [ 1293054 ], "fullName": "Dunfan Lu", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Meta" }, { "questionId": 123918, "answerValue": "Software Engineer" } ] }, { "id": "fe44db8a-c695-465f-b248-d58aba846c72", "firstName": "Eikan", "lastName": "Wang", "bio": "Eikan is a principal engineer from Intel and a DL framework architect having full-stack experience in DL, from various AI applications to framework, library, and DL compiler. He is actively optimizing on torch.compile stack for Intel platforms, including optimizing Inductor C++/OpenMP backend for XEON and Intel XPU Backend for Intel GPU on the top of Triton. In addition, He is deeply involved in the PyTorch community and focuses on PyTorch framework feature design, development, and optimization.", "tagLine": "Intel, Principal Engineer", "profilePicture": "https://cdn.sessionize.com/image/514c-400o400o1-NbxUL3HgNvK3Z1vtPB9b77.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/eikanwang/", "linkType": "LinkedIn" } ], "sessions": [ 1222331 ], "fullName": "Eikan Wang", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Intel" }, { "questionId": 123918, "answerValue": "Principal Engineer" } ] }, { "id": "b7541db4-ccd8-42f8-8d81-53a75ae4e07c", "firstName": "Emily", "lastName": "Chen", "bio": "Emily is a sales/solutions engineer at Akuity, an Argo enthusiast, and a CNCF Ambassador. She is an active contributor to various CNCF projects, and an advocate for diversity and inclusion. She is involved with translation and localization of numerous documentations for open-source projects, including the CNCF Glossary.", "tagLine": "CNCF Ambassador | Sales/Solutions Engineer @ Akuity ", "profilePicture": "https://cdn.sessionize.com/image/f67b-400o400o1-svdKuxWwQqLFkHZ3w5PpvD.jfif", "isTopSpeaker": false, "links": [ { "title": "Twitter", "url": "https://twitter.com/em_chen2", "linkType": "Twitter" }, { "title": "LinkedIn", "url": "https://www.linkedin.com/in/emily-xinyi-chen/", "linkType": "LinkedIn" } ], "sessions": [ 1322749 ], "fullName": "Emily Chen", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Akuity " }, { "questionId": 123918, "answerValue": "Solution Engineer" } ] }, { "id": "a3c542ee-6ee8-44e3-9588-f2f4c1faa06d", "firstName": "Endre", "lastName": "Sara", "bio": "Endre is a Co-Founder of Causely, where he’s building the IT industry’s first causal reasoning. Previously, Endre was VP of Advanced Engineering at Turbonomic. Prior to Turbonomic, Endre was a VP at Goldman Sachs. Endre holds an M.E. in Electrical Engineering from the Technical University of Budapest and a Ph.D. in Electrical Engineering from Stevens Institute of Technology.", "tagLine": "Co-Founder", "profilePicture": "https://cdn.sessionize.com/image/e57f-400o400o1-pLmjErgRwAsQJhJRUBZgC6.JPG", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/endresara/", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://www.causely.ai/", "linkType": "Company_Website" }, { "title": "Sessionize", "url": "https://sessionize.com/endre-sara", "linkType": "Sessionize" } ], "sessions": [ 1219307 ], "fullName": "Endre Sara", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Causely" }, { "questionId": 123918, "answerValue": "Co-Founder" } ] }, { "id": "93d37137-a3ac-42e1-b31c-41d4fbf86253", "firstName": "Feng", "lastName": "Gao", "bio": "Feng Gao is a software engineer and active contributor to the KubeEdge project, an open-source CNCF project that extends Kubernetes to edge computing scenarios. ", "tagLine": "Sony Shanghai China", "profilePicture": "https://cdn.sessionize.com/image/bb25-400o400o1-EEni65uYhSisPdHfgRXqkx.jpg", "isTopSpeaker": false, "links": [ { "title": "Sessionize", "url": "https://sessionize.com/feng-gao", "linkType": "Sessionize" } ], "sessions": [ 1212199 ], "fullName": "Feng Gao", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Sony China" }, { "questionId": 123918, "answerValue": "Software Engineer" } ] }, { "id": "1413a1ce-360d-4c56-8dd5-c820b1cd9aa9", "firstName": "Fengjun", "lastName": "Lyu", "bio": "I'm Mingcheng (明城). Originally from Ningbo, I am currently based in Hangzhou, China. I work as an open-source evangelist at Ant Group and serve on the Project Management Committee (PMC) at the Apache Software Foundation. I actively maintain several open-source projects, including InclusionAI, Dragonfly, and Apache Geaflow.", "tagLine": "Ant Group, Technical Evangelist ", "profilePicture": "https://cdn.sessionize.com/image/f2cb-400o400o1-HGs8bAPipabeAmu7Ph9uiB.png", "isTopSpeaker": false, "links": [ { "title": "X (Twitter)", "url": "https://twitter.com/feelinglucky", "linkType": "Twitter" }, { "title": "Sessionize", "url": "https://sessionize.com/mingcheng", "linkType": "Sessionize" } ], "sessions": [ 1189573 ], "fullName": "Fengjun Lyu", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Ant Group" }, { "questionId": 123918, "answerValue": "Technical Evangelist " } ] }, { "id": "b6340cc0-5435-4d62-8352-7bda98552c72", "firstName": "Fubang", "lastName": "Li", "bio": "Lifubang is a runc Maintainer and LFAPAC Open Source Evangelist, focusing on Linux container runtime security and cloud native threat modeling. He has discovered 4 CVEs related to runc and container security, and has deep experience in upstream container security, including vulnerability triage, threat model design, and secure-by-design practices. He actively contributes to open source security projects and promotes best practices for cloud native security in the Chinese community.", "tagLine": "runc Maintainer, LFAPAC Open Source Evangelist, discoverer of 4 runc/container runtime CVEs", "profilePicture": "https://cdn.sessionize.com/image/aa8b-400o400o1-YrrHk5wsQwyeGGfNKDygd2.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/lifubang/", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://www.acmcoder.com", "linkType": "Company_Website" }, { "title": "Sessionize", "url": "https://sessionize.com/fubang-li", "linkType": "Sessionize" } ], "sessions": [ 1224844 ], "fullName": "Fubang Li", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Acmcoder.com From CIIC" }, { "questionId": 123918, "answerValue": "CTO of acmcoder.com" } ] }, { "id": "87a1caef-272a-4dd5-9390-cd7e06480345", "firstName": "Fupan", "lastName": "Li", "bio": " I worked in Ant Group , which maintained the largest Kata Containers cluster, as an staff engineer, I’m responsible for the implementation and development of the kata container inside Ant, and I am confident that based on my development and land experience on the Ant Group.", "tagLine": "staff engineer", "profilePicture": "https://cdn.sessionize.com/image/c0bc-400o400o1-7V25HJD97ErYTenRwM8Hux.jpg", "isTopSpeaker": false, "links": [ { "title": "Company Website", "url": "https://www.antgroup.com/", "linkType": "Company_Website" } ], "sessions": [ 1309304, 1220661 ], "fullName": "Fupan Li", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Ant Group" }, { "questionId": 123918, "answerValue": "Staff Engineer" } ] }, { "id": "36602fff-5c9a-4b8a-bbfd-16168f1c9a0a", "firstName": "Haibin", "lastName": "Zhang", "bio": "Alibaba Cloud technical expert, currently responsible for application observability-related technology research and development work in the Alibaba Cloud observability team. He has 10 years of work experience in service governance, service mesh and other technology research and development.", "tagLine": "Alibaba Cloud", "profilePicture": "https://cdn.sessionize.com/image/b3ad-400o400o1-wxLLfsmEv9LafkBYNaEW8x.jpg", "isTopSpeaker": false, "links": [ { "title": "Sessionize", "url": "https://sessionize.com/haibin-zhang", "linkType": "Sessionize" } ], "sessions": [ 1219307 ], "fullName": "Haibin Zhang", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Alibaba Cloud" }, { "questionId": 123918, "answerValue": "OpenTelemetry Maintainer & Software Engineer " } ] }, { "id": "6545c39f-4dd2-4489-b230-c723558d0610", "firstName": "HanSol", "lastName": "Park", "bio": "Han Sol Park is a Senior Cloud Native Engineer at Samsung SDS and a former CNCF Ambassador. With nearly a decade of experience in cloud-native technologies like Kubernetes, he specializes in building and scaling complex ecosystems. He currently serves on the team building the underlay cloud architecture for the Samsung Cloud Platform (SCP). Han Sol is dedicated to bridging infrastructure with modern orchestration to deliver robust, production-ready solutions.", "tagLine": "Senior Engineer", "profilePicture": "https://cdn.sessionize.com/image/7ec4-400o400o1-hJHzLXouHNJGbcLZdDBLK1.png", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://kr.linkedin.com/in/hsolpark", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://www.samsungsds.com/", "linkType": "Company_Website" }, { "title": "Sessionize", "url": "https://sessionize.com/hansol-park", "linkType": "Sessionize" } ], "sessions": [ 1224992 ], "fullName": "HanSol Park", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Samsung SDS" }, { "questionId": 123918, "answerValue": "Senior Cloud Native Engineer, Samsung SDS" } ] }, { "id": "a1d5009a-4f79-4103-a478-67eba7b40c43", "firstName": "Hanyue", "lastName": "He", "bio": "Hanyue He is a DevTech Compute Engineer at NVIDIA China. She has expertise in LLM inference. Hanyue earned a bachelor’s degree from Beijing University of Posts and Telecommunications and a master’s degree from Hong Kong University.", "tagLine": "DevTech Compute Engineer, NVIDIA", "profilePicture": "https://cdn.sessionize.com/image/e222-400o400o1-YnzHXZQSdwkkA9bFDKsUSF.jpg", "isTopSpeaker": false, "links": [], "sessions": [ 1224567 ], "fullName": "Hanyue He", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "NVIDIA" }, { "questionId": 123918, "answerValue": "DevTech Compute Engineer" } ] }, { "id": "1351ad0e-c69b-412a-9a35-3505e6398656", "firstName": "Hiu Yeung, Sunny", "lastName": "Chan", "bio": "Sunny Chan is Chief Consultant at TCC, Linux Foundation Authorized Instructor, and elite Golden Kubestronaut with nearly 20 years in enterprise infrastructure and open-source innovation. He leads transformative AI platform initiatives and intelligent infrastructure strategy. Recognized with top industry honors in technical education and consulting, he advances technology ecosystems across Asia-Pacific, nurturing talent and bridging strategic vision with execution.", "tagLine": "Golden Kubestronaut | LFAPAC Evangelist | LFAPAC Education SIG | Principal Consultant", "profilePicture": "https://cdn.sessionize.com/image/1276-400o400o1-YPxNWBurqAYpQ5Ngc5uQJd.png", "isTopSpeaker": false, "links": [ { "title": "Company Website", "url": "https://www.tcc-consulting.com.hk", "linkType": "Company_Website" } ], "sessions": [ 1309304 ], "fullName": "Hiu Yeung, Sunny Chan", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "TCC Consulting Limited" }, { "questionId": 123918, "answerValue": "Golden Kubestronaut | LFAPAC Evangelist | LFAPAC Education SIG" } ] }, { "id": "67f65a78-9fba-498c-84d1-ddf27bac6de3", "firstName": "Hongbing", "lastName": "Zhang", "bio": "Hongbing Zhang is Chief Operating Officer of DaoCloud. He is a veteran in open source areas, he founded IBM China Linux team in 2011 and organized team to make significant contributions in Linux Kernel/openstack/hadoop projects. Now he is focusing on cloud native domain and leading DaoCloud's edge computing tech/product/business team to continue open source contributions.", "tagLine": "Chief Operating Officer Of DaoCloud KubeEdge TSC Member", "profilePicture": "https://cdn.sessionize.com/image/0f08-400o400o1-3MUJiHQwzbmsimQh3U7tCA.png", "isTopSpeaker": false, "links": [ { "title": "Company Website", "url": "https://www.daocloud.io", "linkType": "Company_Website" } ], "sessions": [ 1224863 ], "fullName": "Hongbing Zhang", "categoryItems": [], "questionAnswers": [ { "questionId": 134288, "answerValue": "Maintainer" } ] }, { "id": "1afdc8f0-ebbb-4f55-be52-64bf0815b421", "firstName": "Hongcai", "lastName": "Ren", "bio": "Hongcai Ren(@RainbowMango) is the CNCF Ambassador, who has been working on Kubernetes and other CNCF projects since 2019,\r\nand is the maintainer of the Kubernetes and Karmada projects.", "tagLine": "Huawei, Senior Software Engineer, Karmada", "profilePicture": "https://cdn.sessionize.com/image/8c5d-400o400o1-9kGRVq4e4CZLpXhoHnc3kw.png", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/hongcairen/", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://www.huawei.com/", "linkType": "Company_Website" } ], "sessions": [ 1222021, 1222012 ], "fullName": "Hongcai Ren", "categoryItems": [], "questionAnswers": [ { "questionId": 134288, "answerValue": "Maintainer" } ] }, { "id": "16f68e1f-cee0-4a9c-ae4f-e6dcb34a31db", "firstName": "Hongjun", "lastName": "Zhang", "bio": null, "tagLine": null, "profilePicture": null, "isTopSpeaker": false, "links": [], "sessions": [ 1224095 ], "fullName": "Hongjun Zhang", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "BAAI" }, { "questionId": 123918, "answerValue": "Technical Lead of High-Perf Operator Library" } ] }, { "id": "a50dbbe3-5e92-44ca-9e40-c1b39986cbc7", "firstName": "Horace", "lastName": "Li", "bio": "Horace is currently working as Director of CNCF Asia, supporting China’s cloud native ecosystem and open source projects. Horace is also the Director of OpenInfra Asia. Previously, Horace worked in the Intel Open Source Technology Center for 13 years, as a technical account manager supporting open source community projects engagement in China, including Cloud Computing, DataCenter and Client Computing. Outside of work, Horace likes jogging, swimming, music, movies, fashion and art.", "tagLine": null, "profilePicture": "https://cdn.sessionize.com/image/b185-400o400o1-7yb3qQ4pqgrKC6QdD73NpH.jpg", "isTopSpeaker": true, "links": [], "sessions": [ 1304838, 1304839 ], "fullName": "Horace Li", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "The Linux Foundation" }, { "questionId": 123918, "answerValue": "Director of CNCF Asia, PyTorch Asia, OpenInfra Asia" }, { "questionId": 138874, "answerValue": "https://events.linuxfoundation.org/wp-content/uploads/2022/02/linux-foundation-vert-color.svg" } ] }, { "id": "1892791a-a1af-4fe9-b7c8-ee89de2c76c3", "firstName": "Huabing (Robin)", "lastName": "Zhao", "bio": "Huabing (Robin) Zhao is a software engineer at Tetrate and a CNCF ambassador. He has developed a managed service mesh product on the cloud and assisted a lot of users in deploying Istio service mesh in production. He also founded Aeraki Mesh, a CNCF sandbox project that facilitates non-HTTP protocol management within Istio service mesh. \r\nhttps://zhaohuabing.com", "tagLine": "Engineer, Tetrate", "profilePicture": "https://cdn.sessionize.com/image/2940-400o400o1-DmcSBP9KjQXzQiprM8RGVj.jpg", "isTopSpeaker": false, "links": [ { "title": "Twitter", "url": "https://twitter.com/zhaohuabing", "linkType": "Twitter" }, { "title": "LinkedIn", "url": "https://www.linkedin.com/in/zhaohuabing/", "linkType": "LinkedIn" }, { "title": "Blog", "url": "https://zhaohuabing.com", "linkType": "Blog" }, { "title": "Company Website", "url": "https://tetrate.io", "linkType": "Company_Website" }, { "title": "Sessionize", "url": "https://sessionize.com/huabing-zhao", "linkType": "Sessionize" } ], "sessions": [ 1212758 ], "fullName": "Huabing (Robin) Zhao", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "tetrate.io" }, { "questionId": 123918, "answerValue": "Engineer" } ] }, { "id": "ff2aa64c-9b1b-48fc-b5d9-a85e5d8694c2", "firstName": "Huan", "lastName": "Wei", "bio": "KubeCon Europe speaker, KubeEdge TSC", "tagLine": "Kubernetes, edge computing, high performance computing, etc", "profilePicture": "https://cdn.sessionize.com/image/2a5e-400o400o1-XCEZ8zh98dBaQZM9tsSgXJ.png", "isTopSpeaker": false, "links": [ { "title": "Sessionize", "url": "https://sessionize.com/edonwei", "linkType": "Sessionize" } ], "sessions": [ 1225125 ], "fullName": "Huan Wei", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Harmonycloud" }, { "questionId": 123918, "answerValue": "Senior Technical Director" } ] }, { "id": "bc0746bf-bfbe-44c5-bf98-42dcaaa48ac7", "firstName": "Jack", "lastName": "Khuu", "bio": "Software Engineer @ Meta Superintelligence Labs\r\n\r\nCurrently working on Agentic Kernel Generation and GPUMode\r\n\r\nPreviously Worked on torchchat, torchforge, Executorch\r\nEx-Lecturer @ University of Washington", "tagLine": "Software Engineer on PyTorch Edge @ Meta", "profilePicture": "https://cdn.sessionize.com/image/60df-400o400o1-7tA9UXVLSPHcxniS3HRJgt.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/jack-khuu-uwcse/", "linkType": "LinkedIn" } ], "sessions": [ 1209151 ], "fullName": "Jack Khuu", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Meta" }, { "questionId": 123918, "answerValue": "Software Engineer" } ] }, { "id": "44e11070-13f7-47e1-8ab3-9b2a5ff64d36", "firstName": "Jane", "lastName": "Lyu", "bio": "Jane Lyu is the China Community Manager, overseeing ecosystem operations for CNCF, OpenInfra, and PyTorch. With 20+ years as a Senior Technical Manager at Intel, she specializes in cloud software, Linux systems, and building collaborative open-source ecosystems. Jane has facilitated large-scale cloud technology adoption in China and currently chairs the China Linux Kernel Developers Conference. She is dedicated to bridging local technical practices with the global open-source evolution.", "tagLine": "China Community Manager of the Linux Foundation", "profilePicture": "https://cdn.sessionize.com/image/f434-400o400o1-ptMDi4SspdxKkp43ckjc1x.jpg", "isTopSpeaker": false, "links": [ { "title": "Sessionize", "url": "https://sessionize.com/jane-lyu", "linkType": "Sessionize" } ], "sessions": [ 1309304 ], "fullName": "Jane Lyu", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "The Linux Foundation" }, { "questionId": 123918, "answerValue": "China Community Manager" } ] }, { "id": "bcfc2c5d-dca4-47dc-affa-62de55829f5e", "firstName": "Jane", "lastName": "Xu", "bio": "Hi, I'm Jane! Please don't hesitate to come talk to me about your favorite optimizer, fitting models in GPU memory, how to free C++ extensions from libtorch version, and anything that interests you.", "tagLine": "PyTorch SWE @ Meta", "profilePicture": "https://cdn.sessionize.com/image/b429-400o400o1-hpcMQsDxGtNgLYUERiSDjD.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/jane-yuan-xu/", "linkType": "LinkedIn" }, { "title": "Sessionize", "url": "https://sessionize.com/jane-xu", "linkType": "Sessionize" } ], "sessions": [ 1224946 ], "fullName": "Jane Xu", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Meta Platforms Inc" }, { "questionId": 123918, "answerValue": "PyTorch SWE @ Meta" } ] }, { "id": "4dbc9a9e-ef1f-4e4d-b0d1-0c8442cf1b71", "firstName": "Jared", "lastName": "Tan", "bio": "Jared Tan is a Sr. Software Engineer at DaoCloud responsible for the Observability Platform, which includes contributions to the OpenTelemetry project and with a passion for observability and helping users start their observability journey. He has participated in several well-known open source projects, he is steering committee member of Apache Skywalking project, and is a triager OpenTelemetry Collector and approver for OTel Helm Charts project now.", "tagLine": "DaoCloud, Observability Engineer, OpenTelemetry member", "profilePicture": "https://cdn.sessionize.com/image/f553-400o400o1-V4zR1RkLvy5QPzVuoHkMKk.png", "isTopSpeaker": false, "links": [ { "title": "Blog", "url": "https://github.com/JaredTan95/", "linkType": "Blog" }, { "title": "Company Website", "url": "https://www.daocloud.io/", "linkType": "Company_Website" }, { "title": "DaoCloud", "url": "https://www.daocloud.io/", "linkType": "Other" }, { "title": "Sessionize", "url": "https://sessionize.com/jian-tan", "linkType": "Sessionize" } ], "sessions": [ 1223860 ], "fullName": "Jared Tan", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "DaoCloud" }, { "questionId": 123918, "answerValue": "Software Engineer" } ] }, { "id": "6ac086c7-2c70-4b54-96ad-2e816c314f31", "firstName": "Jason", "lastName": "Ansel", "bio": "Jason Ansel is a Research Scientist at Meta AI and a technical lead for PyTorch compilers. He started the TorchDynamo and TorchInductor projects, which bring flexible graph capture and a high performance compiler to PyTorch 2. He received a Ph.D. from MIT and has over 15 years of experience in machine learning, compilers, and programming languages.", "tagLine": "PyTorch Compilers @ Meta", "profilePicture": "https://cdn.sessionize.com/image/0063-400o400o1-wNnFdS3APbELud6XyYdi2K.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/jansel/", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://jasonansel.com", "linkType": "Company_Website" } ], "sessions": [ 1293054 ], "fullName": "Jason Ansel", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Meta" }, { "questionId": 123918, "answerValue": "Software Engineer" } ] }, { "id": "4f75f000-e518-49a6-b3a1-0d79346c5220", "firstName": "Jeff", "lastName": "Hammond", "bio": "Jeff Hammond is a Distinguished Engineer in the NCCL team at NVIDIA focused on user education and research outreach. His background is in parallel application and algorithm development, open-source software, and supercomputing architecture. Jeff has made significant contributions to open standards, especially MPI.", "tagLine": "Distinguished Engineer, NVIDIA", "profilePicture": "https://cdn.sessionize.com/image/5fe2-400o400o1-DsLRdAaYZgCXzocv9HmECk.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "http://linkedin.com/in/jeffhammond", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://www.nvidia.com/", "linkType": "Company_Website" }, { "title": "Sessionize", "url": "https://sessionize.com/jeff-hammond", "linkType": "Sessionize" } ], "sessions": [ 1224567 ], "fullName": "Jeff Hammond", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "NVIDIA" }, { "questionId": 123918, "answerValue": "Distinguished Engineer" } ] }, { "id": "9936935c-dfa6-49c1-91bd-13281ab40901", "firstName": "Jiahang", "lastName": "Xu", "bio": "Jiahang Xu is a System Architect at China Merchants Bank. He has over 15 years of unique cross-domain experience working in telecom, automotive, financial industry, startup as a co-founder, and KubeVela maintainer. He's mainly focused on cloud-native application technology practice in recent years. He is seasoned in the application management system and is responsible for building large-scale financial platform engineering for China Merchants Bank.", "tagLine": "KubeVela Maintainer, LFAPAC Open Source Evangelist, Speaker for KubeCon, Apsara Con, System Architect at China Merchants Bank, ", "profilePicture": "https://cdn.sessionize.com/image/3cf1-400o400o1-DC3mTKi2EkgyomsisQPt2C.png", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/jiahang-xu-89122152/", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "http://www.cmbchina.com/", "linkType": "Company_Website" }, { "title": "Sessionize", "url": "https://sessionize.com/jiahang-xu", "linkType": "Sessionize" } ], "sessions": [ 1225364 ], "fullName": "Jiahang Xu", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "China Merchants Bank" }, { "questionId": 123918, "answerValue": "Senior Software Engineer" } ] }, { "id": "b39c90e5-f927-43b6-aba9-3f38853fd0cf", "firstName": "Jiamu", "lastName": "Liu", "bio": "Jiamu LIU is an AI Researcher at the China Mobile Research Institute, specializing in AI for Science (AI4S). Her research focuses on AI/ML, deep learning, neural networks, and NLP, driving the next generation of machine learning applications for scientific advancement.", "tagLine": "China Mobile Research Institute", "profilePicture": "https://cdn.sessionize.com/image/9d57-400o400o1-G92Pc1GHSULvLSG1LZiXZi.jpg", "isTopSpeaker": false, "links": [ { "title": "Sessionize", "url": "https://sessionize.com/jiamu-liu", "linkType": "Sessionize" } ], "sessions": [ 1224322 ], "fullName": "Jiamu Liu", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "China Mobile Research Institute" }, { "questionId": 123918, "answerValue": "AI Researcher" } ] }, { "id": "6e6ea1fd-e7fb-4e69-9958-aca0438036e6", "firstName": "Jian", "lastName": "Qiu", "bio": "Qiu Jian is an active contributor to the Kubernetes community, primarily focused on SIG-Multicluster and SIG-Auth, and a\r\n maintainer of the Open Cluster Management project. He is the author of KEP-5284 Constrained Impersonation. Qiu Jian has spoken at KubeCon China twice and is passionate about building\r\n secure, scalable multi-cluster infrastructure.", "tagLine": "Independent", "profilePicture": "https://cdn.sessionize.com/image/f414-400o400o1-Wm9VxV4JRLvtJLcM8gP8g3.png", "isTopSpeaker": false, "links": [ { "title": "Sessionize", "url": "https://sessionize.com/jian-qiu", "linkType": "Sessionize" } ], "sessions": [ 1218125 ], "fullName": "Jian Qiu", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Independent" }, { "questionId": 123918, "answerValue": "Freelancer" } ] }, { "id": "06f510fe-e4e0-4482-8675-88959f9a2282", "firstName": "JIanxiang", "lastName": "Sun", "bio": "Jianxiang Sun is a Senior Cloud-Native Engineer at Horizon Robotics, where he specializes in scaling workflow systems and scheduling for massive autonomous driving pipelines. He previously built machine learning platforms at Kuaishou and led scientific computing scheduling at DP Technology. An expert in Kubernetes efficiency and Go runtime optimization, Jianxiang Sun is dedicated to resolving bottlenecks in extreme-scale task orchestration and AI infrastructure.", "tagLine": "Horizon Robotics, Software Engineer", "profilePicture": "https://cdn.sessionize.com/image/9845-400o400o1-PkE9EGgX92zCx9QR7gRR94.jpg", "isTopSpeaker": false, "links": [ { "title": "Sessionize", "url": "https://sessionize.com/jian-xiang-sun", "linkType": "Sessionize" } ], "sessions": [ 1225136 ], "fullName": "JIanxiang Sun", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Horizon Robotics" }, { "questionId": 123918, "answerValue": "Software Engineer" } ] }, { "id": "66e515d5-3e82-4f74-864e-5dc52a0a7985", "firstName": "Jiawei", "lastName": "Li", "bio": "7+ years experience on open source, worked on OpenStack development.\r\n\r\nCurrently work on contributing AI ecosystem.\r\n- PyTorch Maintainer of PrivateUse1 Module\r\n- ONNXRuntime Ascend Support Author", "tagLine": "Staff Engineer - Huawei Technologies Co., Ltd.", "profilePicture": "https://cdn.sessionize.com/image/395e-400o400o1-PVUeDSELXPAbGNag4roSN9.jpg", "isTopSpeaker": false, "links": [], "sessions": [ 1222276 ], "fullName": "Jiawei Li", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Huawei" }, { "questionId": 123918, "answerValue": "Staff Engineer" } ] }, { "id": "c28d10bb-0c49-4886-acca-d8dd5de44b37", "firstName": "Jiayu", "lastName": "Sun", "bio": "Nvidia, DevTech Engineer", "tagLine": "Nvidia, DevTech Engineer", "profilePicture": "https://cdn.sessionize.com/image/9802-400o400o1-srWNoMMQxaNbjWcvUHfj6M.png", "isTopSpeaker": false, "links": [], "sessions": [ 1210254 ], "fullName": "Jiayu Sun", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "NVIDIA" }, { "questionId": 123918, "answerValue": "Software Engineer" } ] }, { "id": "f1bebe17-344f-4716-9b12-901b184b967f", "firstName": "Jingbin", "lastName": "Zhang", "bio": "Designed and developed Unified Cache Manager.\r\nRecent speeches at open source events:\r\nAugust 2025, vLLM Meetup in Shenzhen shared \"Practices of UCM Long-Sequence Inference Optimization\" \r\nSeptember 2025, GOSIM in Hangzhou shared \"UCM Inference Acceleration Solution\" ", "tagLine": "Chief Expert of Huawei Inference Storage", "profilePicture": "https://cdn.sessionize.com/image/abb1-400o400o1-wnzj5R9p6hgKdE2q7duYxM.jpg", "isTopSpeaker": false, "links": [], "sessions": [ 1223545 ], "fullName": "Jingbin Zhang", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Huawei" }, { "questionId": 123918, "answerValue": "Chief Expert of Inference Storage" } ] }, { "id": "71d5f45d-6d99-4e32-a0fc-54bc07b69532", "firstName": "Jingwei", "lastName": "Huang", "bio": "CI/CD and large-scale test automation specialist. TorchNPU maintainer. Led mcore architecture transformation for MindSpore Transformers and integrated MindSpore with vLLM, with a focus on inference optimization at scale.", "tagLine": "Staff Engineer - Huawei Technologies Co., Ltd.", "profilePicture": "https://cdn.sessionize.com/image/18d0-400o400o1-HwTpuVnFVSgPPQ8GFX2C4w.jpg", "isTopSpeaker": false, "links": [ { "title": "Sessionize", "url": "https://sessionize.com/jingwei-huang", "linkType": "Sessionize" } ], "sessions": [ 1222276 ], "fullName": "Jingwei Huang", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Huawei" }, { "questionId": 123918, "answerValue": "Staff Engineer" } ] }, { "id": "7805aad6-9d9c-4ade-b72f-cc935d18679d", "firstName": "Jionghang", "lastName": "Cai", "bio": "Senior Technical Expert in AI Infrastructure at Alibaba, specializing in large-scale AI compute, training and inference infrastructure, and AI data platforms.", "tagLine": "Senior Technical Expert, AI Infrastructure, Alibaba", "profilePicture": "https://cdn.sessionize.com/image/b55b-400o400o1-JQ1nsowcsWpVGjj2BYhsG5.jpg", "isTopSpeaker": false, "links": [], "sessions": [ 1317079 ], "fullName": "Jionghang Cai", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Alibaba" }, { "questionId": 123918, "answerValue": "Senior Technical Expert, AI Infrastructure" } ] }, { "id": "af7b6fb1-4084-4706-a495-3fc1793d97f2", "firstName": "Jonathan", "lastName": "Bryce", "bio": "Jonathan Bryce is the Executive Director of Cloud & Infrastructure at the Linux Foundation, where he leads both the Cloud Native Computing Foundation (CNCF) and the OpenInfra Foundation—two of the largest and most influential open source communities in the world. With over 15 years of open source leadership, Jonathan co-founded the OpenStack project, guiding its evolution into a global ecosystem that supports a multi-billion-dollar market and serves as the de facto standard for open cloud infrastructure. Before his open source career, Jonathan spent over a decade in data center infrastructure and co-founded one of the industry’s earliest cloud computing companies, which grew into a billion-dollar business following its acquisition by Rackspace. Passionate about open collaboration, infrastructure innovation, and empowering users, contributors, and ecosystems worldwide, Jonathan frequently speaks at international conferences on open source strategy, infrastructure modernization, cloud-native technologies, and open source AI initiatives.", "tagLine": null, "profilePicture": "https://cdn.sessionize.com/image/edd2-400o400o1-s8jTYPb6kZeCjL37FMXQye.jpg", "isTopSpeaker": true, "links": [], "sessions": [ 1304838, 1304839 ], "fullName": "Jonathan Bryce", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "The Linux Foundation" }, { "questionId": 123918, "answerValue": "Executive Director, Cloud and Infrastructure" }, { "questionId": 138874, "answerValue": "https://events.linuxfoundation.org/wp-content/uploads/2022/02/linux-foundation-vert-color.svg" } ] }, { "id": "4627a7ee-a0d9-4001-acfd-8cdabc022077", "firstName": "Jun", "lastName": "Duan", "bio": "Jun Duan is a Staff Research Scientist at IBM Thomas J. Watson Research Center, working on cloud computing and AI systems. He contributes to Kubernetes, vLLM, and llm-d across infrastructure and AI workloads. He is a maintainer of KubeStellar (CNCF Sandbox project) focused on multi-cluster and edge-native orchestration. He has published in systems and networking venues and served on the program committee for KubeCon.", "tagLine": "Staff Research Scientist, IBM", "profilePicture": "https://cdn.sessionize.com/image/edbd-400o400o1-hNu3G6HNMHoYpQbP5CwAVW.jpg", "isTopSpeaker": false, "links": [ { "title": "Sessionize", "url": "https://sessionize.com/jun-duan", "linkType": "Sessionize" } ], "sessions": [ 1218005 ], "fullName": "Jun Duan", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "IBM" }, { "questionId": 123918, "answerValue": "Staff Research Scientist" } ] }, { "id": "d896475c-5e85-4b88-829a-153e2912b1ca", "firstName": "Jun", "lastName": "Zheng", "bio": null, "tagLine": null, "profilePicture": null, "isTopSpeaker": false, "links": [], "sessions": [ 1317065 ], "fullName": "Jun Zheng", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "China Merchants Bank" }, { "questionId": 123918, "answerValue": "Inference Optimization Engineer" } ] }, { "id": "378341b9-6f6f-420a-aff7-63128c669668", "firstName": "Kaichao", "lastName": "You", "bio": "Co-founder and Chief Scientist @ Inferact Inc.\r\nCore Maintainer of vLLM project.", "tagLine": "Co-founder and Chief Scientist @ Inferact Inc.", "profilePicture": "https://cdn.sessionize.com/image/770a-400o400o1-QCysYYCkuVRXeVzuyNcGxy.jpg", "isTopSpeaker": false, "links": [], "sessions": [ 1311164 ], "fullName": "Kaichao You", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Inferact" }, { "questionId": 123918, "answerValue": "Co-founder and Chief Scientist @ Inferact Inc." } ] }, { "id": "0b25d981-368c-4648-8ab6-e480d86991b4", "firstName": "Kailun", "lastName": "Qin", "bio": "Kailun Qin is a maintainer of Gramine LibOS (https://github.com/gramineproject), leading the development and innovation for its SGX and TDX backends at Intel Product Assurance and Security Group.\r\n\r\nKailun has published research in ACM (CCS, Middleware) and IEEE (SRDS, CLOUD, TDSC) and given 10+ talks at conferences including KubeCon, Cloud Native SecurityCon, IstioCon and OpenInfra Summit. He has also contributed to RISC-V CoVE and OCI specifications to help shape the future of secure systems.", "tagLine": "Intel Corporation, Staff Cloud Software Engineer", "profilePicture": "https://cdn.sessionize.com/image/b79f-400o400o1-UTidNNzBRo3P6RePp2kNk7.png", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/kailun-qin", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://www.intel.com", "linkType": "Company_Website" }, { "title": "Sessionize", "url": "https://sessionize.com/kailun-qin", "linkType": "Sessionize" } ], "sessions": [ 1221758 ], "fullName": "Kailun Qin", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Intel Corporation" }, { "questionId": 123918, "answerValue": "Staff Cloud Software Engineer" } ] }, { "id": "b33fc27e-2735-4035-a83f-39b7790e9e44", "firstName": "Kaiming", "lastName": "Cheng", "bio": "Kaiming Cheng is current a research scientist at Meta SuperIntelligence Labs, PyTorch Team working on KernelAgent.\r\n\r\nHe received his Ph.D. from the Paul G. Allen School of Computer Science & Engineering at the University of Washington, and earned his bachelor’s degree from the University of Virginia. ", "tagLine": "Meta SuperIntelligence Labs - PyTorch Team| Research Scientist", "profilePicture": "https://cdn.sessionize.com/image/90e3-400o400o1-EfxWvqSECUsm2Ua1CXq3TC.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/kaiming-cheng-b7732aa3/", "linkType": "LinkedIn" }, { "title": "Sessionize", "url": "https://sessionize.com/kaiming-cheng", "linkType": "Sessionize" } ], "sessions": [ 1209151 ], "fullName": "Kaiming Cheng", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Meta" }, { "questionId": 123918, "answerValue": "Research Scientist" } ] }, { "id": "27e5f808-79cf-4a54-9512-d572400f9e67", "firstName": "Kang", "lastName": "Zhang", "bio": "He is a senior solution architect. He specializes in AI workload scheduling and optimization, with deep expertise in distributed training and Kubernetes-based deployment of large language models (LLMs). ", "tagLine": "Nvidia, Senior Solution Architect", "profilePicture": "https://cdn.sessionize.com/image/01c3-400o400o1-Dy3iQbydNBSrJVVsUKdfk.jpg", "isTopSpeaker": false, "links": [], "sessions": [ 1223308 ], "fullName": "Kang Zhang", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "NVIDIA" }, { "questionId": 123918, "answerValue": "Senior Solution Architect" } ] }, { "id": "73b45691-d9d8-4744-8755-f9b7fac55079", "firstName": "Kangsub", "lastName": "Song", "bio": "Kangsub Song is a Senior Enginner at Samsung SDS. He joined in 2012, and large-scale Java and C backend systems until 2014.\r\nFrom 2015 he led Docker, container, and Kubernetes projects, createing CI/CD pipelines and automating clusters.\r\nNow he designs and builds SCP, Samsung's private-cloud platform, forcusing on multi-tenant security, cost efficieny. Additionally, he presents at industry conferences and publishes articles on cloudnative best practices.", "tagLine": "Samsung SDS / Senior Engineer", "profilePicture": "https://cdn.sessionize.com/image/71b1-400o400o1-RGNCCwxdacuP7JfQPTi3TY.png", "isTopSpeaker": false, "links": [], "sessions": [ 1224992 ], "fullName": "Kangsub Song", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Samsung SDS" }, { "questionId": 123918, "answerValue": "Senior Engineer" } ] }, { "id": "29eabf0a-4372-4f76-b8b9-4deb70ff2a79", "firstName": "Kay", "lastName": "Yan", "bio": "Principal Software Engineer at DaoCloud, focused on Kubernetes-native AI infrastructure, distributed inference, and production LLM serving. He has 15+ years of experience in Kubernetes and CNCF ecosystems, and contributes to Kubespray, LeaderWorkerSet, containerd/nerdctl, LLM-D, Kubean, Spiderpool, HwameiStor, and Merbridge. He holds 23 Chinese patents and 9 U.S. patents. Previously, he was a Senior Technologist at DELL/EMC and received the EMC Innovation CTO Award.", "tagLine": "Maintainer of kubespray,containerd/nerdctl,LWS(LeaderWorkerSet) and LLM-D, Software Engineer in DaoCloud", "profilePicture": "https://cdn.sessionize.com/image/19a3-400o400o1-UCRHDHyuXPzS77V6zuJ2sD.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/yankay/", "linkType": "LinkedIn" }, { "title": "Blog", "url": "http://www.yankay.io/", "linkType": "Blog" }, { "title": "Company Website", "url": "http://www.daocloud.io/", "linkType": "Company_Website" }, { "title": "Github", "url": "https://github.com/yankay", "linkType": "Other" }, { "title": "Sessionize", "url": "https://sessionize.com/kay-yan", "linkType": "Sessionize" } ], "sessions": [ 1222146 ], "fullName": "Kay Yan", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "DaoCloud" }, { "questionId": 123918, "answerValue": "Principal Software Engineer" } ] }, { "id": "ff78d921-13c1-4c10-822d-c81dd4168e22", "firstName": "Ke", "lastName": "Bao", "bio": "Core Developer and Maintainer of SGLang, and Member of Technical Staff at RadixArk. Key contributor to day-0 support and performance optimization for frontier LLMs. Builds and optimizes core runtime components for high-performance LLM inference and serving.", "tagLine": "SGLang Core Developer", "profilePicture": "https://cdn.sessionize.com/image/81f8-400o400o1-M3S9UzFWo5y9Ljxea27nVi.png", "isTopSpeaker": false, "links": [ { "title": "Company Website", "url": "https://www.radixark.com/", "linkType": "Company_Website" } ], "sessions": [ 1325851 ], "fullName": "Ke Bao", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "RadixArk" }, { "questionId": 123918, "answerValue": "Member of Technical Staff" } ] }, { "id": "f9b10346-995d-4c20-a3a4-fdabef35fb13", "firstName": "Ke", "lastName": "Wen", "bio": "Ke Wen works on distributed features, including Symmetric Memory, multi-GPU kernels, Expert Parallelism, inference, pipelining and graph analysis.", "tagLine": "NVIDIA, Principal Software Architect, PyTorch Maintainer", "profilePicture": "https://cdn.sessionize.com/image/27d4-400o400o1-VFxj2YPNHGXGVgAXidGDer.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://linkedin.com/in/ke-wen-601a35b3", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://pytorch.org/", "linkType": "Company_Website" }, { "title": "Sessionize", "url": "https://sessionize.com/ke-wen", "linkType": "Sessionize" } ], "sessions": [ 1224567 ], "fullName": "Ke Wen", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "NVIDIA" }, { "questionId": 123918, "answerValue": "Principal Software Architect" } ] }, { "id": "26415447-f9a1-4909-a6df-6695199e792f", "firstName": "Kebe", "lastName": "Liu", "bio": "DaoCloud’s inference engine architect specializes in cloud-native technologies and optimizing the performance of inference engines. He was also a member of the Istio steering committee and the creator of the Merbridge project.", "tagLine": "DaoCloud, Architect", "profilePicture": "https://cdn.sessionize.com/image/98e9-400o400o1-LuJRFQ3FTi91i4nuYA2xDX.png", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/kebe-liu-27044713a/", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://daocloud.io/", "linkType": "Company_Website" }, { "title": "Sessionize", "url": "https://sessionize.com/kebe", "linkType": "Sessionize" } ], "sessions": [ 1220647 ], "fullName": "Kebe Liu", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "DaoCloud" }, { "questionId": 123918, "answerValue": "Architect" } ] }, { "id": "a4305a38-d462-482b-9e4a-536eb74ec63c", "firstName": "Kevin", "lastName": "Dubois", "bio": "Kevin Dubois is often featured as a (keynote) speaker at conferences around the world, where he shares his passion and knowledge about developer experience, open source, cloud native development and Java. He is also an author, java Champion, and an accomplished software architect and platform engineer. Kevin currently works as a Senior Principal Developer Advocate at IBM, and is also Technical Lead for the CNCF Developer Experience Technical Advisory Group.\r\n", "tagLine": "Senior Principal Developer Advocate at IBM", "profilePicture": "https://cdn.sessionize.com/image/7df3-400o400o1-WoJXMV1WNg5WJ88aFuYTku.jpg", "isTopSpeaker": false, "links": [ { "title": "X (Twitter)", "url": "https://twitter.com/kevindubois", "linkType": "Twitter" }, { "title": "LinkedIn", "url": "https://www.linkedin.com/in/kevindubois/", "linkType": "LinkedIn" }, { "title": "Blog", "url": "https://www.kevindubois.com", "linkType": "Blog" }, { "title": "Company Website", "url": "https://ibm.com", "linkType": "Company_Website" }, { "title": "Youtube", "url": "https://youtube.com/@thekevindubois", "linkType": "Other" }, { "title": "Mastodon", "url": "https://mastodon.social/@kevindubois", "linkType": "Other" }, { "title": "Bluesky", "url": "https://bsky.app/profile/kevindubois.com", "linkType": "Other" }, { "title": "Sessionize", "url": "https://sessionize.com/kevindubois", "linkType": "Sessionize" } ], "sessions": [ 1180062 ], "fullName": "Kevin Dubois", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "IBM" }, { "questionId": 123918, "answerValue": "Sr Principal Developer Advocate" } ] }, { "id": "1b814ba0-bc81-4245-880a-097b1e7e3b75", "firstName": "Kevin", "lastName": "Wang", "bio": "Kevin Wang has been an outstanding contributor in the CNCF community since its beginning and is the leader of the cloud native open source team at Huawei. Kevin has contributed critical enhancements to Kubernetes, led the incubation of the KubeEdge, Volcano, Karmada projects in CNCF, actively participated in the TOC, and played a critical role in building and supporting the growth of CNCF technologies and the CNCF community in the globe.", "tagLine": "CNCF TOC Vice Chair, Ambassador, Kubernetes emeritus Maintainer, Founder and Maintainer of multiple CNCF projects, Lead of Cloud Native Open Source Team at Huawei", "profilePicture": "https://cdn.sessionize.com/image/9c71-400o400o1-h2y88cTaoPHJW7cQM2PScp.png", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/kevin-wangzefeng/", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://huaweicloud.com", "linkType": "Company_Website" }, { "title": "Sessionize", "url": "https://sessionize.com/kevin-wang", "linkType": "Sessionize" } ], "sessions": [ 1224671 ], "fullName": "Kevin Wang", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Huawei" }, { "questionId": 123918, "answerValue": "Vice Chair, CNCF TOC | Technical Expert at Huawei" } ] }, { "id": "2d6077f4-3da7-43b6-bd0d-eea01cef0486", "firstName": "Laura", "lastName": "Wang", "bio": "Laura Wang leads KernelAgent at Meta's Super Intelligence Lab, an agentic system for autonomous GPU and TPU kernel generation. Her team landed the first agent-generated kernel in vLLM, with CuTeDSL output beating expert baselines on Blackwell and Triton kernels hitting 1.56× over torch.compile on H100. She works across CUDA/PTX, CuTeDSL, Triton, and Pallas on deep agent architectures for production-quality code generation.", "tagLine": "MSL-Pytorch | Software Engineer Tech Lead", "profilePicture": "https://cdn.sessionize.com/image/8335-400o400o1-nsFbkF43q7ZabtXDePtRGh.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/leywang", "linkType": "LinkedIn" } ], "sessions": [ 1209151 ], "fullName": "Laura Wang", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Meta" }, { "questionId": 123918, "answerValue": "Software Engineer" } ] }, { "id": "f7834264-1b08-4f5a-874f-e0654a5a3016", "firstName": "Leon", "lastName": "Nunes", "bio": "Leon Nunes is a Senior Customer Success Engineer working around Cloud-native technologies, Agents, Agent-gateways and Service meshes, he's also active in various Open-Source communities and Cloud native communities and loves talking about his varied interests in tech.", "tagLine": "Senior Customer Success Engineer|Platform advocate", "profilePicture": "https://cdn.sessionize.com/image/3144-400o400o1-GTei3LJ4ohQfddz7LjrKFP.jpg", "isTopSpeaker": false, "links": [ { "title": "X (Twitter)", "url": "https://twitter.com/mediocreDevops", "linkType": "Twitter" }, { "title": "LinkedIn", "url": "https://linkedin.com/in/leon-nunes", "linkType": "LinkedIn" }, { "title": "Blog", "url": "https://dev.to/mediocreDevops", "linkType": "Blog" }, { "title": "Company Website", "url": "https://solo.io", "linkType": "Company_Website" }, { "title": "Sessionize", "url": "https://sessionize.com/leon-nunes", "linkType": "Sessionize" } ], "sessions": [ 1319502 ], "fullName": "Leon Nunes", "categoryItems": [], "questionAnswers": [] }, { "id": "716a62ee-7c24-4d67-a609-1ba94bc8b7eb", "firstName": "Liang ", "lastName": "Zhang", "bio": "Liang joined Huawei in 2012 and boasts profound experience in IT product sales, strategic planning, marketing operations and ecosystem development. He is dedicated to advancing the open source initiative of Ascend CANN and building the AI industrial ecosystem, and driving collaborative co development between Ascend and global open source communities.", "tagLine": null, "profilePicture": "https://cdn.sessionize.com/image/601b-400o400o1-LqCZw5j5UdWCdtAReBnrvR.jpg", "isTopSpeaker": false, "links": [], "sessions": [ 1317073 ], "fullName": "Liang Zhang", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Huawei" }, { "questionId": 123918, "answerValue": "Vice President of Huawei Ascend Computing Product Line" } ] }, { "id": "25cc1809-2d06-4971-a2c9-bfdfa4538d5b", "firstName": "Liyi", "lastName": "Huang", "bio": "senior solution architect @isovalent.com", "tagLine": "isovalent, solution architect, k8s networking", "profilePicture": "https://cdn.sessionize.com/image/81ee-400o400o1-suo1WmaSR1i5QYDqbFtjXv.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/liyi-huang-a7776238/", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://isovalent.com/", "linkType": "Company_Website" }, { "title": "Sessionize", "url": "https://sessionize.com/liyi-huang", "linkType": "Sessionize" } ], "sessions": [ 1227863 ], "fullName": "Liyi Huang", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Isovalent (Cisco)" }, { "questionId": 123918, "answerValue": "solution architect" } ] }, { "id": "d1e7e65a-bf95-44a0-97fe-440ba8a5c310", "firstName": "Mark", "lastName": "Collier", "bio": "Mark Collier is a longtime open-source strategist and the current General Manager, AI & Infrastructure at the Linux Foundation. He co-founded both the OpenStack project and the OpenInfra Foundation, guiding a small NASA–Rackspace collaboration into one of the most active open-source communities in history igniting a multi-billion-dollar cloud market. At the Linux Foundation he now focuses on the next wave of open technology—AI—working with companies, researchers, and developers to build a vendor-neutral stack that keeps intelligence infrastructure accessible to all. A regular keynote speaker and commentator, Mark translates deep technical shifts into clear strategic insight and for rallying diverse communities to deliver the future of computing in the open.", "tagLine": null, "profilePicture": "https://cdn.sessionize.com/image/33a7-400o400o1-TC91doDWQ3AXGAqke1rudU.jpg", "isTopSpeaker": true, "links": [], "sessions": [ 1312046 ], "fullName": "Mark Collier", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "PyTorch Foundation" }, { "questionId": 123918, "answerValue": "Executive Director" }, { "questionId": 138874, "answerValue": "https://events.linuxfoundation.org/wp-content/uploads/2026/08/PyTorchFLogo_Horiz_fullColor-1-scaled.png" } ] }, { "id": "8010363b-334b-482e-825d-ee6acf5c7cf2", "firstName": "Maxim", "lastName": "Nazarenko", "bio": "Maxim Nazarenko is a Senior Software Development Engineer at Amazon with 10+ years of experience building scalable software and distributed systems at Amazon, CloudKitchens, Booking.com, and Microsoft. Maxim combines research-driven approach to problems enhanced by hands-on experience in solving customers' painpoints and delivering results at scale.\r\n\r\nBefore transitioning into software engineering, Maxim earned a PhD in semiconductor physics working with low-dimensional nanostructures. ", "tagLine": "Saving business problem through sofware", "profilePicture": "https://cdn.sessionize.com/image/f892-400o400o1-Ux6Y3Hcf6pzASnHwrQfVQb.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/maxim-nazarenko-7b657888/", "linkType": "LinkedIn" } ], "sessions": [ 1224867 ], "fullName": "Maxim Nazarenko", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Amazon" }, { "questionId": 123918, "answerValue": "Senior Software Engineer" } ] }, { "id": "edb71874-107d-4c2b-a50f-72fe29a26d23", "firstName": "Mengqing", "lastName": "Cao", "bio": "I am an AI Infrastructure Engineer at Huawei, vLLM Ascend maintainer. My work focuses on the intersection of LLM inference efficiency, specifically deep-diving into KV Cache scheduling and holistic performance optimization.", "tagLine": "Staff Engineer - Huawei Technologies Co., Ltd.", "profilePicture": "https://cdn.sessionize.com/image/92ea-400o400o1-U2ArC8px8CHHeS38PaHzUH.jpg", "isTopSpeaker": false, "links": [], "sessions": [ 1222763 ], "fullName": "Mengqing Cao", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Huawei" }, { "questionId": 123918, "answerValue": "Staff Engineer" } ] }, { "id": "b5c46ce4-07b8-450f-a296-f7f37df30671", "firstName": "Mengsi", "lastName": "Lyu", "bio": "Developer of vllm-plugin-FL, with 7 years of experience in accelerating large models.", "tagLine": "Beijing Academy of Artificial Intelligence, vllm-plugin-FL developer", "profilePicture": "https://cdn.sessionize.com/image/b529-400o400o1-TC6zMJdL6ov5B7iM5kXMXU.jpg", "isTopSpeaker": false, "links": [], "sessions": [ 1224095 ], "fullName": "Mengsi Lyu", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "BAAI" }, { "questionId": 123918, "answerValue": "Head of AI Framework Inference" } ] }, { "id": "32272b85-b0be-43ab-8267-b6263d4c2b35", "firstName": "Mengxuan", "lastName": "Li", "bio": "Maintainer of HAMi community\r\nApprover of volcano community\r\n\r\nResponsible for the development of gpu virtualization mechanism on HAMi and volcano. Working on providing device-sharing on all heterogeneous devices and improving device utilization on clusters.\r\n\r\n", "tagLine": "Dynamia.ai", "profilePicture": "https://cdn.sessionize.com/image/d155-400o400o1-7TcY8uN17m2GkTZrMZryVj.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/mengxuan-li-5862a9314", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://www.4paradigm.com/", "linkType": "Company_Website" }, { "title": "Sessionize", "url": "https://sessionize.com/li-mengxuan", "linkType": "Sessionize" } ], "sessions": [ 1313959 ], "fullName": "Mengxuan Li", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Dynamia.ai" }, { "questionId": 123918, "answerValue": "Co-founder and CTO" } ] }, { "id": "12dfc79a-ac6a-463e-8229-6429ce9a29e5", "firstName": "Miley", "lastName": "Fu", "bio": "Miley is the co-chair and keynote speaker for KubeCon+Open Source Summit and AI Dev China 2024. She works on WasmEdge runtime under Linux Foundation as the founding member for over 6 years. She talks at KubeCon, KCD Beijing+Shenzhen, CloudDay Italy, DevRelCon, Open Source Summit Japan, AWS User Group, Global AI Note, KubeDay Singapore etc. Miley writes tech content and organizes dev events, including KCDs, MCP and AI Agent Meetups, GOSIM AI PARIS, Wasm track at FOSDEM, in EU, the US, Asia, etc.", "tagLine": "CNCF Ambassador, Founding member of open source runtime WasmEdge", "profilePicture": "https://cdn.sessionize.com/image/89e3-400o400o1-aEenhyjXuvVhxsBLwPP3Br.jpg", "isTopSpeaker": false, "links": [ { "title": "X (Twitter)", "url": "https://twitter.com/mileyfu", "linkType": "Twitter" }, { "title": "LinkedIn", "url": "https://www.linkedin.com/in/mileyfu/", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://www.olares.com", "linkType": "Company_Website" }, { "title": "China Stay", "url": "https://chinastay.org/", "linkType": "Other" }, { "title": "Sessionize", "url": "https://sessionize.com/mileyfu", "linkType": "Sessionize" } ], "sessions": [ 1260073, 1260075 ], "fullName": "Miley Fu", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Olares " }, { "questionId": 123918, "answerValue": "DevRel" } ] }, { "id": "d5414f6a-5825-45e1-aac1-801ff7ad9baa", "firstName": "Monty", "lastName": "Taylor", "bio": "Monty was part of the team who launched the OpenStack project. He defined the workflows and built the initial developer infrastructure that became Zuul and OpenDev. Many years ago he worked for MySQL, and then was a core dev on Drizzle. These days his hacking is focused on WanderTracks, a travel planning and logging system, and marine automation systems.", "tagLine": "Co-founder of Zuul, OpenDev and OpenStack", "profilePicture": "https://cdn.sessionize.com/image/5b25-400o400o1-XffCYP9UGEaiM2pfSgmqy1.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/montytaylor/", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://wandertracks.com/", "linkType": "Company_Website" }, { "title": "Sessionize", "url": "https://sessionize.com/monty-taylor", "linkType": "Sessionize" } ], "sessions": [ 1221033 ], "fullName": "Monty Taylor", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "WanderTracks" }, { "questionId": 123918, "answerValue": "Founder" } ] }, { "id": "72e83064-d18e-4278-9bc6-2a3e43549995", "firstName": "Morgan", "lastName": "Foster", "bio": "Morgan Foster works on distributed inference and platform primitives for agentic systems. She is a maintainer of the Kagenti project, a framework-neutral middleware for managing agent workloads on Kubernetes. ", "tagLine": "Red Hat, Senior Principal Software Engineer", "profilePicture": "https://cdn.sessionize.com/image/4d63-400o400o1-TJqDkPZ62qsaBNKfKZipBR.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/morgan-f-3151a7170/", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://next.redhat.com", "linkType": "Company_Website" } ], "sessions": [ 1177779 ], "fullName": "Morgan Foster", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Red Hat" }, { "questionId": 123918, "answerValue": "Senior Principal Software Engineer" } ] }, { "id": "0992ddcd-389f-4343-9151-68dc8c1c9165", "firstName": "Muhammad", "lastName": "Hamza", "bio": "Hamza is a telecommunications architect and open-source advocate with over five years of experience working with the European Telecommunications Standards Institute (ETSI) to shape the future of Network Functions Virtualisation (NFV). Working at the intersection of formal standardisation and hands-on implementation, Hamza is trying to bridging the gap between traditional telco requirements and the cloud-native ecosystem.", "tagLine": "NTT DOCOMO", "profilePicture": "https://cdn.sessionize.com/image/25a8-400o400o1-DC1hTntSJ1VmoGcfpAhJGE.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/mhamza007", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://docomolab-euro.com/", "linkType": "Company_Website" }, { "title": "Sessionize", "url": "https://sessionize.com/muhammad-hamza", "linkType": "Sessionize" } ], "sessions": [ 1224932 ], "fullName": "Muhammad Hamza", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "DOCOMO Euro-Labs" }, { "questionId": 123918, "answerValue": "Senior Network Architect" } ] }, { "id": "7b380a7c-a033-417e-af64-6741a72e400e", "firstName": "Murphy", "lastName": "Chen", "bio": "Murphy is a Senior Software Engineer at Daocloud where he works in an observability team to build a Multi-cluster observability platform.\r\nHe is a triager of openTelemetry collector contrib, and he is also the codeonwer of clickhouse exporter and spanconnector.", "tagLine": "Daocloud, Senior Software Engineer, triager of openTelemetry collector contrib", "profilePicture": "https://cdn.sessionize.com/image/2d89-400o400o1-nrWXaTqkFKDdJDoWGvvmCY.png", "isTopSpeaker": false, "links": [ { "title": "Company Website", "url": "https://www.daocloud.io", "linkType": "Company_Website" }, { "title": "Sessionize", "url": "https://sessionize.com/murphy-chen", "linkType": "Sessionize" } ], "sessions": [ 1223860 ], "fullName": "Murphy Chen", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "DaoCloud" }, { "questionId": 123918, "answerValue": "Software Engineer" } ] }, { "id": "9566ce0e-7307-4fd9-a90f-d63a1c50131c", "firstName": "Nibir", "lastName": "Bora", "bio": "Nibir is a member of the Core Infrastructure team at a City Storage Systems, where he is responsible for the company's Kubernetes infrastructure running 100s of clusters globally.", "tagLine": "Core Infrastructure at a City Storage Systems", "profilePicture": "https://cdn.sessionize.com/image/538c-400o400o1-Th6qedvfgnigkn98NZ36Ft.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/nibirbora/", "linkType": "LinkedIn" }, { "title": "Sessionize", "url": "https://sessionize.com/nibir-bora", "linkType": "Sessionize" } ], "sessions": [ 1224867 ], "fullName": "Nibir Bora", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Clean Compute" }, { "questionId": 123918, "answerValue": "Engineering Manager" } ] }, { "id": "50e09a13-2711-4c71-b987-611717466667", "firstName": "Nicole", "lastName": "Li", "bio": "Production LLM Inference\r\nOpen Source Contributor: Istio, llm-d, vLLM, GAIE, LWS, etc.", "tagLine": "DaoCloud, Cloud Native & AI Inference Engineer", "profilePicture": "https://cdn.sessionize.com/image/8979-400o400o1-3Fb8US91Zq86VVJpYnCWUw.png", "isTopSpeaker": false, "links": [ { "title": "X (Twitter)", "url": "https://x.com/NicoleLiHui", "linkType": "Twitter" }, { "title": "Company Website", "url": "https://www.daocloud.io/", "linkType": "Company_Website" }, { "title": "Github", "url": "https://github.com/nicole-lihui", "linkType": "Other" }, { "title": "Sessionize", "url": "https://sessionize.com/nicole-li", "linkType": "Sessionize" } ], "sessions": [ 1223860, 1220647 ], "fullName": "Nicole Li", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "DaoCloud" }, { "questionId": 123918, "answerValue": "Cloud Native & AI Inference Engineer" } ] }, { "id": "0d5dfadd-1092-4508-8d50-36e18bf685de", "firstName": "Norwin", "lastName": "Schnyder", "bio": "Norwin Schnyder is a Cloud Software Engineer at Airlock and a Lecturer at the Eastern Switzerland University of Applied Sciences and the University of Applied Sciences and Arts Northwestern Switzerland. With over six years of experience in Kubernetes and various contributions to the Gateway API, he focuses on building secure, cloud-native ecosystems by leveraging the power of Kubernetes operators. Outside of his work, Norwin is a passionate sailor and has experience using Helm at sea.", "tagLine": "Senior Cloud Software Engineer, Airlock", "profilePicture": "https://cdn.sessionize.com/image/0ac2-400o400o1-DhnMxGSjBa253G8XQsueEG.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/norwin-schnyder/airlock", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://www.airlock.com/", "linkType": "Company_Website" }, { "title": "Sessionize", "url": "https://sessionize.com/snorwin", "linkType": "Sessionize" } ], "sessions": [ 1212758 ], "fullName": "Norwin Schnyder", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Airlock" }, { "questionId": 123918, "answerValue": "Senior Cloud Software Engineer" } ] }, { "id": "dd481bd7-e625-4e56-a76b-4baa54d457a5", "firstName": "Paco", "lastName": "Xu", "bio": "Paco is a member of Kubernetes Steering Committee, and co-chair of KubeCon+CloudNativeCon China 2024.\r\n\r\nPaco is a kubeadm maintainer and an active kubernetes contributor. His interest includes SIG-Node, LTS and CoreDNS.\r\n\r\nHe is the leader of the open-source team in DaoCloud. He organized KCD Chengdu 2022 and KCS China 2023, and speaked in KCD Shanghai, KubeCon EU 2023 & 2024 & 2025, and KubeCon China 2021, 2023, 2024 & 2025. He is also LFAPAC Evangelist.", "tagLine": "DaoCloud, Lead of open source team.", "profilePicture": "https://cdn.sessionize.com/image/9acd-400o400o1-SvGVWdN7HRTMDfXf9JHXBn.jpg", "isTopSpeaker": false, "links": [ { "title": "X (Twitter)", "url": "https://twitter.com/xu_paco", "linkType": "Twitter" }, { "title": "LinkedIn", "url": "https://www.linkedin.com/in/pacoxu2020/", "linkType": "LinkedIn" }, { "title": "Blog", "url": "https://pacoxu.wordpress.com/", "linkType": "Blog" }, { "title": "Company Website", "url": "https://www.daocloud.io/", "linkType": "Company_Website" }, { "title": "Github", "url": "https://github.com/pacoxu", "linkType": "Other" }, { "title": "DaoCloud", "url": "https://github.com/DaoCloud", "linkType": "Other" }, { "title": "🦋", "url": "https://bsky.app/profile/pacoxu.bsky.social", "linkType": "Other" }, { "title": "Sessionize", "url": "https://sessionize.com/pacoxu", "linkType": "Sessionize" } ], "sessions": [ 1223308 ], "fullName": "Paco Xu", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "DaoCloud" }, { "questionId": 123918, "answerValue": "Open Source Team Lead" } ] }, { "id": "aaac253c-cd23-4c13-acfb-f65c76175bc6", "firstName": "Peng", "lastName": "Li", "bio": "Peng Li is a technical expert at Alibaba Cloud. He works on Alibaba Cloud Container Service for Kubernetes , focusing on Knative and Kubernetes.", "tagLine": "Alibaba Cloud Technical Expert, Knative Contributor", "profilePicture": "https://cdn.sessionize.com/image/80dd-400o400o1-wmq4794qUjPPjYADFYvxJP.jpg", "isTopSpeaker": false, "links": [], "sessions": [ 1215557 ], "fullName": "Peng Li", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Alibaba Cloud" }, { "questionId": 123918, "answerValue": "Software Engineer" } ] }, { "id": "38662798-6e4b-4de9-96b6-639fb6021de1", "firstName": "Pengfei", "lastName": "Ni", "bio": "Pengfei Ni is an AI engineering practitioner specializing in the integration and implementation of AI agent operating frameworks, AI infrastructure on Kubernetes, and R&D infrastructure. He is dedicated to seamlessly incorporating AI into the daily workflows of software development, R&D collaboration, and infrastructure operations.\r\n\r\nHe is also a columnist for Geek Time and a Kubernetes project maintainer, with over ten years of experience in cloud-native and large-scale distributed systems. ", "tagLine": "Principal Software Engineer, Microsoft", "profilePicture": "https://cdn.sessionize.com/image/2ed4-400o400o1-wBzPS4D47Kux9M865qM2LG.JPG", "isTopSpeaker": false, "links": [ { "title": "Sessionize", "url": "https://sessionize.com/pengfeini", "linkType": "Sessionize" } ], "sessions": [ 1218010 ], "fullName": "Pengfei Ni", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Microsoft" }, { "questionId": 123918, "answerValue": "Principal Software Engineer" } ] }, { "id": "aa1d25a7-7dfe-4fa4-ae67-59c835e5dbae", "firstName": "Peter", "lastName": "Pan", "bio": "- DaoCloud Software Engineering VP\r\n\r\n- Regular KubeCon \"Program Committee\" : 2023 EU, 2024 HK, 2024 India, 2025 EU\r\n- Regular KubeCon Speaker: 2023 SH, 2024 EU, 2024 HK\r\n- CNCF wg-AI (AI Working-Group) member + CNAI white-paper co-author\r\n- Contributor of L.W.S & vLLM\r\n- Maintainer of below CNCF projects : cloudtty, kubean, hwameistor\r\n- GithubID: panpan0000", "tagLine": "Cloud-Native Developer , Open Source Enthusiast", "profilePicture": "https://cdn.sessionize.com/image/0912-400o400o1-Nfn5HBVUPvsW2RrkqW2WGx.png", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://cn.linkedin.com/in/peter-pan-83424073", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://www.daocloud.io/", "linkType": "Company_Website" }, { "title": "Sessionize", "url": "https://sessionize.com/peter-pan", "linkType": "Sessionize" } ], "sessions": [ 1220735 ], "fullName": "Peter Pan", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "DaoCloud" }, { "questionId": 123918, "answerValue": "VP of R&D Engineering" } ] }, { "id": "0150521f-3d02-4425-a1a1-707febc575e9", "firstName": "Professor Lu", "lastName": "Shouqun", "bio": null, "tagLine": null, "profilePicture": "https://cdn.sessionize.com/image/3622-400o400o1-ai3meu5J16LKHBDuv5Hi4v.png", "isTopSpeaker": false, "links": [], "sessions": [ 1308499 ], "fullName": "Professor Lu Shouqun", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "China OSS" }, { "questionId": 123918, "answerValue": "Honorary Chairman" } ] }, { "id": "03dbf5de-879a-4c7c-b9c0-46d68383c559", "firstName": "Quoc Dat", "lastName": "Le", "bio": "With years of experience designing, building, and operating large-scale cloud infrastructure systems based on OpenStack and Kubernetes technologies, my work involves developing cloud platforms that meet demanding performance and reliability requirements, includes infrastructure architecture, container orchestration, and especially networking in cloud environments, with the goal of bridging the gap between mainstream cloud platforms and specialized systems such as telecommunications.\r\n", "tagLine": "Cloud Principal Specialist, Viettel", "profilePicture": "https://cdn.sessionize.com/image/d33c-400o400o1-JopWUYL9A36wuS5Wn1zyA4.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/datle01/", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://viettel.com.vn/", "linkType": "Company_Website" } ], "sessions": [ 1219415 ], "fullName": "Quoc Dat Le", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Viettel" }, { "questionId": 123918, "answerValue": "Cloud Principal Specialist" } ] }, { "id": "ee1a3539-1e0f-442e-a18c-910cb38e1818", "firstName": "Rui", "lastName": "Chen", "bio": "Rui Chen is a Senior Infrastructure Engineer at Justworks and a maintainer of Atlantis, Homebrew, tflint, and awesome-terraform. He has contributed across the Terraform, OpenTofu, Kubernetes, and Bazel ecosystems, and has served on the OpenTofu Day Program Committee. He is interested in open source infrastructure, platform engineering, and practical cloud native automation.", "tagLine": "Engineer", "profilePicture": "https://cdn.sessionize.com/image/6d30-400o400o1-sHBpBtL6VaBcxq2A3S8b7z.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/chenrui333/", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://www.justworks.com", "linkType": "Company_Website" }, { "title": "Sessionize", "url": "https://sessionize.com/chenrui-na", "linkType": "Sessionize" } ], "sessions": [ 1218468 ], "fullName": "Rui Chen", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Justworks" }, { "questionId": 123918, "answerValue": "Senior Infrastructure Engineer" }, { "questionId": 134288, "answerValue": "Maintainer" } ] }, { "id": "82706269-934e-470b-a08c-ac8e33709d7b", "firstName": "Ryan", "lastName": "Lee", "bio": "Ryan serves as Head of Developer Relations at MiniMax. \r\nHe leads MiniMax’s overall open‑source strategy for foundation models and drives global developer ecosystem construction. \r\nHis responsibilities span open‑source releases of LLM models, multimodal models, community operations and developer outreach. He focuses on lowering barriers for developers to adopt MiniMax models and building a vibrant open‑source AI community.", "tagLine": "Head of Devrel, MiniMax", "profilePicture": "https://cdn.sessionize.com/image/a3d2-400o400o1-4gYbUm8vGczERkgBxu9wZf.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/ryanlee-dev/", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://www.minimax.io/", "linkType": "Company_Website" } ], "sessions": [ 1313048 ], "fullName": "Ryan Lee", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "MiniMax" }, { "questionId": 123918, "answerValue": "Head of Devrel" } ] }, { "id": "1ca6dcf8-473e-4737-a075-baabb268dab6", "firstName": "Sean", "lastName": "Chen", "bio": "Xiaohong “Sean” Chen is a Principal Software Engineer at Red Hat. His current work focuses on integrating Helion into vLLM for production LLM inference. Previously at AWS AI, Sean was a tech lead for Amazon Bedrock, building multi-node model hosting, multi-LoRA inference infrastructure, and model optimization features. Sean holds a Ph.D. in Applied Mathematics and has published research in numerical methods and scientific computing.", "tagLine": "Red Hat, Principal Software Engineer", "profilePicture": "https://cdn.sessionize.com/image/6674-400o400o1-DsoMgNsyvYH5d1nuw61Ps1.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/xiaohong/", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://www.redhat.com", "linkType": "Company_Website" } ], "sessions": [ 1214156 ], "fullName": "Sean Chen", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Red Hat" }, { "questionId": 123918, "answerValue": "Principal Software Engineer" } ] }, { "id": "e051c049-14f7-480f-8edb-c02677855aa3", "firstName": "Seongsoo", "lastName": "Cho", "bio": "Seongsoo Cho is an OpenStack engineer and led the OpenInfra Korea User Group from 2020 to 2025. He is currently working with the OpenStack i18n SIG on migrating to a new translation platform. He actively mentors aspiring OpenStack developers and engineers in Korea and is a passionate advocate for OpenStack.", "tagLine": "OpenStack i18n SIG, Maintainer, OpenInfra Korea User Group Organizer", "profilePicture": "https://cdn.sessionize.com/image/a665-400o400o1-aGwoJFSty5YaY6JpWyRmkG.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/seongsoocho/", "linkType": "LinkedIn" } ], "sessions": [ 1225189 ], "fullName": "Seongsoo Cho", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "NHN Cloud" }, { "questionId": 123918, "answerValue": "System Engineer" } ] }, { "id": "b14e2c8e-b7da-4835-9a2c-7c9705ac94dd", "firstName": "Sergey", "lastName": "Pronin", "bio": "Sergey is a product leader focusing on delivering robust open-source solutions. After having consistently contributed to the growth of Percona and its product open source culture as a VP of Product, he is now leading the OpenEverest open source database platform. Digital nomad, who enjoys spreading the word about data on Kubernetes. ", "tagLine": "Solanica, Founder", "profilePicture": "https://cdn.sessionize.com/image/d990-400o400o1-KuSdUZhLgERutzHRLKPYLu.jpg", "isTopSpeaker": false, "links": [ { "title": "X (Twitter)", "url": "https://twitter.com/sergeypronin", "linkType": "Twitter" }, { "title": "LinkedIn", "url": "https://www.linkedin.com/in/sergeypronin/", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://solanica.io", "linkType": "Company_Website" }, { "title": "Sessionize", "url": "https://sessionize.com/sergeypronin", "linkType": "Sessionize" } ], "sessions": [ 1223701 ], "fullName": "Sergey Pronin", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Solanica" }, { "questionId": 123918, "answerValue": "Founder" }, { "questionId": 134288, "answerValue": "Maintainer" } ] }, { "id": "c2ed46cc-5d32-4933-a395-18af73da5d4f", "firstName": "Shendi", "lastName": "Wang", "bio": null, "tagLine": null, "profilePicture": null, "isTopSpeaker": false, "links": [], "sessions": [ 1313956 ], "fullName": "Shendi Wang", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Huawei" }, { "questionId": 123918, "answerValue": "Director, Ascend AI Framework Ecosystem" } ] }, { "id": "975d742e-38fd-4b5f-a244-90b146c587a3", "firstName": "Shuangkun", "lastName": "Tian", "bio": "ShuangKun Tian is a software engineer at Alibaba Cloud, specializing in Scheduling, Elasticity, Workflow Orchestration and Performance tuning. He is one of the maintainer of the CNCF graduated project Argo Workflows and has extensive experience in building MLOps platforms and Data Pipelines on a large scale.", "tagLine": "Alibaba Cloud Technical Expert", "profilePicture": "https://cdn.sessionize.com/image/cf5f-400o400o1-35M6chxWQVs5B8d6fbbKEM.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/shuangkun-tian-89152a351", "linkType": "LinkedIn" }, { "title": "Sessionize", "url": "https://sessionize.com/shaungkun-tian", "linkType": "Sessionize" } ], "sessions": [ 1225136, 1219807 ], "fullName": "Shuangkun Tian", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Alibaba Cloud" }, { "questionId": 123918, "answerValue": "Alibaba Cloud Technical Expert" } ] }, { "id": "b67a8b89-5f45-4650-9456-658b42f8800e", "firstName": "Steve", "lastName": "Baker", "bio": "Steve Baker is a Principal Software Engineer at Red Hat, where he has spent over 14 years working on OpenStack and open infrastructure. He is currently focused on bare metal provisioning with OpenStack Ironic. He has been a core contributor to multiple OpenStack projects spanning the full lifecycle of the platform including OpenStack Heat, TripleO, and Kubernetes-based OpenStack operators. Steve has presented at multiple OpenStack Summits and is based in Wellington, New Zealand.", "tagLine": "Principal Software Engineer, Red Hat", "profilePicture": "https://cdn.sessionize.com/image/e051-400o400o1-GKBPM6NR2CuDzEk1Pu7zR.jpg", "isTopSpeaker": false, "links": [ { "title": "Company Website", "url": "https://redhat.com", "linkType": "Company_Website" } ], "sessions": [ 1223326 ], "fullName": "Steve Baker", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Red Hat" }, { "questionId": 123918, "answerValue": "Principal Software Engineer" } ] }, { "id": "5bccdb2d-dac4-4b1f-a0b5-a7355828e71d", "firstName": "Tan", "lastName": "Pei Xiang", "bio": " Tan Peixiang is a Senior AI Infrastructure Engineer at China Merchants Bank and Lead Maintainer of Twinkle. He specializes in distributed systems and decoupling algorithmic semantics from infrastructure. He pioneered the \"Training-as-a-Service\" (TaaS) model at CMB for multi-tenant LLM post-training on heterogeneous clusters. Through Twinkle, he aims to democratize complex RL (e.g., GRPO) via modular and scalable architectures for the developer community.", "tagLine": "Senior AI Infrastructure Engineer, China Merchants Bank | Lead Maintainer of Twinkle", "profilePicture": "https://cdn.sessionize.com/image/ee73-400o400o1-YXn7cS3UkQzNPZkzFtFqxg.png", "isTopSpeaker": false, "links": [ { "title": "Sessionize", "url": "https://sessionize.com/tan-pei-xiang", "linkType": "Sessionize" } ], "sessions": [ 1317065, 1224575 ], "fullName": "Tan Pei Xiang", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "China Merchants Bank" }, { "questionId": 123918, "answerValue": "Senior AI Infrastructure Engineer, China Merchants Bank | Lead Maintainer of Twinkle" } ] }, { "id": "74274563-3c5e-4b6b-9da6-0ab62d49fab4", "firstName": "Teng", "lastName": "Su", "bio": "Teng Su is the Chief Expert of Huawei's AI Framework Technology Team, specializing in large-scale Ascend cluster training infrastructure. With extensive experience conducting LLM training experiments across tens of thousands of Ascend NPUs, he is currently leading the development of HyperParallel—a parallel acceleration library designed for Ascend SuperPoDs.", "tagLine": "Chief AI Framework Technical Expert, Huawei", "profilePicture": "https://cdn.sessionize.com/image/7b0e-400o400o1-sVjevTgM4LKtoCvzhz3TaV.jpg", "isTopSpeaker": true, "links": [], "sessions": [ 1313956 ], "fullName": "Teng Su", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Huawei" }, { "questionId": 123918, "answerValue": "Chief AI Framework Technical Expert" }, { "questionId": 138874, "answerValue": "https://events.linuxfoundation.org/wp-content/uploads/2020/02/Huawei.svg" } ] }, { "id": "66f55a66-d9bd-4877-8c69-4a2186b74bef", "firstName": "Thierry", "lastName": "Carrez", "bio": "Thierry Carrez is the General Manager of the OpenInfra Foundation. A systems engineer by trade, he was involved in the inception of the OpenStack project, and still contributes to its governance and release management.\r\n\r\nWith a personal interest in the intersection of sociology and technology, Thierry spoke about open innovation and open source at various conferences around the world, including OSCON, LinuxCon, and FOSDEM. He currently serves as Secretary and Director for the Open Source Initiative. He was recognized as a Python Software Foundation fellow in 2012, and previously worked as the Technical lead for Ubuntu Server at Canonical, an operational manager for the Gentoo Linux Security Team, and an IT manager in various companies.", "tagLine": null, "profilePicture": "https://cdn.sessionize.com/image/54a4-400o400o1-9X8yFZwFnEHNTDqHsTuAeE.jpg", "isTopSpeaker": true, "links": [], "sessions": [ 1312069 ], "fullName": "Thierry Carrez", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "OpenInfra Foundation" }, { "questionId": 123918, "answerValue": "General Manager" }, { "questionId": 138874, "answerValue": "https://events.linuxfoundation.org/wp-content/uploads/2023/04/OpenInfraFoundation-logo-RGB-horiz.svg" } ] }, { "id": "42620d7f-2a95-4c85-917b-6baf8e290916", "firstName": "Tiejun", "lastName": "Chen", "bio": "Tiejun Chen is a seasoned technical leader. He is one RISC-V Ambassador of RISC-V International, and ever worked at several tech companies such as VMware, Intel, Wind River Systems and so on, involved in - cloud native, RISC-V, edge computing, ML/AI, etc. He joined to share many presentations at PyTorch Conference 2025, RISC-V Summit 2019 & 2022 & 2023, AI.Dev NA 2023, kubecon China 2021, 2024 & 2025, Kube Edge day Europe 2022, OSS North America 2017 & 2018 & 2024, etc.", "tagLine": "Tech lead", "profilePicture": "https://cdn.sessionize.com/image/3729-400o400o1-M8gh9amh3DW9zWE7fN3U5c.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/tiejun-chen-41a6654b/", "linkType": "LinkedIn" } ], "sessions": [ 1184985 ], "fullName": "Tiejun Chen", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "RISC-V International " }, { "questionId": 123918, "answerValue": "Ambassador" } ] }, { "id": "b244f2b8-cb23-4be6-948d-58010ee7e019", "firstName": "Tingjin", "lastName": "Ye", "bio": "Focus on architecting and scaling the cloud-native ecosystem and AI/ML infrastructure for ANTA Group.", "tagLine": "ANTA Group,Staff Infrastructure Engineer,Cloud Native & AI/ML Platform", "profilePicture": "https://cdn.sessionize.com/image/2569-400o400o1-s4StXQpeAKTeqQYvPX26h2.png", "isTopSpeaker": false, "links": [ { "title": "Company Website", "url": "https://www.anta.com", "linkType": "Company_Website" }, { "title": "Sessionize", "url": "https://sessionize.com/tingjin-ye", "linkType": "Sessionize" } ], "sessions": [ 1227863 ], "fullName": "Tingjin Ye", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "ANTA Sports Products Limited" }, { "questionId": 123918, "answerValue": "Staff Infrastructure Engineer" } ] }, { "id": "dde599d6-57ee-4fbc-9a07-ae71ff0c3c61", "firstName": "Tristan", "lastName": "Rice", "bio": "Software engineer working on PyTorch Distributed and large scale training.", "tagLine": "Meta - PyTorch", "profilePicture": "https://cdn.sessionize.com/image/073c-400o400o1-MqqJhe63E4uV9wjch6zfv6.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://linkedin.com/in/d4l3k", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://fn.lc", "linkType": "Company_Website" } ], "sessions": [ 1223450 ], "fullName": "Tristan Rice", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Meta" }, { "questionId": 123918, "answerValue": "Software engineer" } ] }, { "id": "0f17947b-02a6-404d-8878-3fe915a9639e", "firstName": "Trong Vinh", "lastName": "Nguyen", "bio": "- Deputy Director at \"Plarforms and Digital Transformation Center\" - Viettel Networks - Viettel Group\r\n- Organizer of Vietnam Open Infrastructure Community (formerly VietOpenStack)\r\n- Organizer of Cloud Native Vietnam Community (including Hanoi and Ho Chi Minh chapter)\r\n- Member of OpenInfra Asia Advisory Board\r\n- Organized lots of OpenInfra technical meetup, OpenInfra Days in Vietnam, Cloud Native Days Vietnam", "tagLine": "Deputy Director, Platforms and Digital Transformation Center, Viettel Networks, Viettel Group. Organizer of Vietnam Open Infrastructure and Cloud Native Vietnam Community.", "profilePicture": "https://cdn.sessionize.com/image/c6a0-400o400o1-aEacaB5D3pwpUKcrbLWKFf.png", "isTopSpeaker": true, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/tovin07/", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://viettel.com.vn/en/", "linkType": "Company_Website" }, { "title": "Sessionize", "url": "https://sessionize.com/trong-vinh-nguyen", "linkType": "Sessionize" } ], "sessions": [ 1308448, 1219415 ], "fullName": "Trong Vinh Nguyen", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Viettel" }, { "questionId": 123918, "answerValue": "Deputy Director" }, { "questionId": 138874, "answerValue": "https://events.linuxfoundation.org/wp-content/uploads/2026/08/Viettel_logo_2021.svg" } ] }, { "id": "0ed2ceb9-abe1-4de0-ba1d-64da3a79ad5f", "firstName": "Vincent", "lastName": "Caldeira", "bio": "Vincent Caldeira, Red Hat APAC CTO and Industry Visiting Scholar at Columbia University, drives tech strategy and emerging engineering. A Top 10 APAC CTO (2023) with 20+ years in finance IT, he is an authority on open source, cloud-native technologies and AI. Vincent holds leadership roles across the Linux Foundation (FINOS, LF AI & Data) and GSF, championing AI safety, sustainability, and digital sovereignty.", "tagLine": "Leading Open Source Technology Innovation for a Sustainable Future", "profilePicture": "https://cdn.sessionize.com/image/fcc9-400o400o1-PT9QpE3suViKyBFoABzuE8.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/caldeirav/", "linkType": "LinkedIn" }, { "title": "Blog", "url": "https://thinkinginsidethebox.ai/", "linkType": "Blog" }, { "title": "Company Website", "url": "https://www.redhat.com/", "linkType": "Company_Website" }, { "title": "Sessionize", "url": "https://sessionize.com/vincent-caldeira", "linkType": "Sessionize" } ], "sessions": [ 1177779 ], "fullName": "Vincent Caldeira", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Red Hat" }, { "questionId": 123918, "answerValue": "CTO APAC" } ] }, { "id": "5ecf3700-16a6-4832-b340-e7b3b86beca9", "firstName": "Walter", "lastName": "Duan", "bio": "Walter is the Head of Cloud & Infrastructure at INTSIG and founder of Chaterm.ai, an open-source AI-native terminal for cloud infrastructure. With over 10 years of SRE experience, he has led the design and operation of large-scale systems. Through his work at an AWS Advanced Partner, he has built deep hands-on expertise and frequently shares his insights at conferences such as AWS Summit and QCon.", "tagLine": "Head of Cloud & Infrastructure at INTSIG, Founder of the open-source project Chaterm.ai", "profilePicture": "https://cdn.sessionize.com/image/9159-400o400o1-NrM51UHLLMcx7Mf4KUn1C3.png", "isTopSpeaker": false, "links": [ { "title": "Company Website", "url": "https://www.intsig.com/en", "linkType": "Company_Website" } ], "sessions": [ 1221808 ], "fullName": "Walter Duan", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "intsig information co. ltd" }, { "questionId": 123918, "answerValue": "Head of Cloud & Infrastructure " } ] }, { "id": "6b9fccdc-3e14-44f5-9d72-34cd95db79b9", "firstName": "Wang", "lastName": "Cong", "bio": "Huawei Data Storage – KVCache Acceleration Expert for LLM, with research focuses on High-Performance Computing (HPC), LLM Inference Acceleration, and AI Infrastructure. After joining Huawei, primarily responsible for incubating and designing cutting-edge storage technologies. Currently dedicated to the incubation of Huawei’s UCM (Unified Cache Manager) centered around KVCache, leading UCM architecture design, key feature development, and open-source community operations.", "tagLine": "Software Engineer, Huawei Technologies Co.", "profilePicture": "https://cdn.sessionize.com/image/807e-400o400o1-fbYMAcmvmqGru9YLDRbUxn.jpg", "isTopSpeaker": false, "links": [ { "title": "Sessionize", "url": "https://sessionize.com/wang-cong", "linkType": "Sessionize" } ], "sessions": [ 1223545 ], "fullName": "Wang Cong", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Huawei" }, { "questionId": 123918, "answerValue": "Software Engineer" } ] }, { "id": "75419934-433e-4c5f-8e7e-277a2c12f39f", "firstName": "Wei", "lastName": "Li", "bio": "Wei Li is a Principal Software Engineer at Cambricon, where he leads a team focused on PyTorch and the AI software stack for Cambricon platforms. His team drives hardware backend enablement, framework integration, and upstream contributions to the open‑source community, with a strong focus on compilers, distributed training, and high‑performance systems for large‑scale AI workloads.", "tagLine": "Principal Software Engineer at Cambricon", "profilePicture": "https://cdn.sessionize.com/image/9c8a-400o400o1-TpeYhJvAYgWpGzQxqFVbDq.jpg", "isTopSpeaker": false, "links": [], "sessions": [ 1313965 ], "fullName": "Wei Li", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Cambricon" }, { "questionId": 123918, "answerValue": "Principal Software Engineer" } ] }, { "id": "2509fe84-b256-432f-accd-e786bef7734c", "firstName": "Wei", "lastName": "Wang", "bio": "Wang Wei, currently a professor at the School of Data Science and Engineering, East China Normal University, X-lab open laboratory founder. Outstanding member of CCF, member of CCF Open Source Development Committee, person in charge of open source governance series standards of China Electronics Standardization Institute, member of Mulan Open Source Community Technical Committee, director of Open Source Club. His research interests is open source development ecology.", "tagLine": "Professor of East China Normal University", "profilePicture": "https://cdn.sessionize.com/image/c9d3-400o400o1-ppTduZcHdbdZoE2Ns9x5h7.jpg", "isTopSpeaker": false, "links": [], "sessions": [ 1225138 ], "fullName": "Wei Wang", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "East China Normal University" }, { "questionId": 123918, "answerValue": "Professor" } ] }, { "id": "0b3bfa50-14ae-4363-bf39-d130fe277c1e", "firstName": "Weizhou", "lastName": "Lan", "bio": "I currently serve as a Senior Tech Lead at DaoCloud, with over 15 years of engineering experience, including 8 years of Kubernetes expertise. My technical focus spans networking, eBPF, service mesh, AI, and more. I have delivered 5 KubeCon talks and 3 KCD talks, and have served on the KubeCon program committee 4 times. I have incubated 6 cloud-native open-source projects, including one CNCF Sandbox. Over the past three years, I ranked 28th globally in CNCF developer activity contributions.", "tagLine": "daocloud network cloud-native", "profilePicture": "https://cdn.sessionize.com/image/4f19-400o400o1-HVP6qxBNagLUZVAbLnTLdG.png", "isTopSpeaker": false, "links": [ { "title": "Sessionize", "url": "https://sessionize.com/lan-wei-zhou", "linkType": "Sessionize" } ], "sessions": [ 1224457 ], "fullName": "Weizhou Lan", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "DaoCloud" }, { "questionId": 123918, "answerValue": "senior tech lead" } ] }, { "id": "13b40d9b-27d7-42f1-a074-21e7e84a442f", "firstName": "Wenbo", "lastName": "Qi", "bio": "Wenbo Qi is a software engineer at Ant Group and a maintainer of Dragonfly. Dedicated to contributing to the open-source community, he believes that fear springs from ignorance.", "tagLine": "Dragonfly Maintainer, ModelPack Maintainer", "profilePicture": "https://cdn.sessionize.com/image/7d6d-400o400o1-92Ffhb7Df26TBe9Zk2gVqk.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/gaius-qi-%E6%88%9A%E6%96%87%E5%8D%9A-3238911a3/", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://www.antgroup.com/", "linkType": "Company_Website" }, { "title": "Sessionize", "url": "https://sessionize.com/wenbo-qi", "linkType": "Sessionize" } ], "sessions": [ 1189573, 1174542 ], "fullName": "Wenbo Qi", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Ant Group" }, { "questionId": 123918, "answerValue": "Software Engineer" } ] }, { "id": "83fc1a01-1dc2-438c-8e99-b036ba9b67be", "firstName": "Wenjia", "lastName": "Zhang", "bio": "Wenjia Zhang is an Engineer Manager at Google, working on Google Kubernetes Engine and Google Distributed Cloud. She is an active contributor for both Kubernetes and etcd open source projects.", "tagLine": "Google", "profilePicture": "https://cdn.sessionize.com/image/5a0f-400o400o1-wVXs4uRxY1a7xPvSnQTVLK.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/zhangwenjia/", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://about.google/", "linkType": "Company_Website" } ], "sessions": [ 1210983 ], "fullName": "Wenjia Zhang", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Google" }, { "questionId": 123918, "answerValue": "Engineering Manager" } ] }, { "id": "32e5b704-5b3e-4a4c-ac27-9cbcbd6da7a9", "firstName": "Wenjie", "lastName": "Bao", "bio": "As an Software Engineer within the Wireless Research Institute at ZTE Corporation, I specialize in building and optimizing next-generation cloud infrastructure. My core expertise lies at the intersection of OpenStack, AI Infrastructure, and high-performance networking via RDMA.", "tagLine": "ZTE, Software Engineer", "profilePicture": "https://cdn.sessionize.com/image/c37f-400o400o1-Kbvj545Lv8N7RzxeuDAuh9.jpg", "isTopSpeaker": false, "links": [], "sessions": [ 1224832 ], "fullName": "Wenjie Bao", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "ZTE" }, { "questionId": 123918, "answerValue": "Software Engineer" } ] }, { "id": "3fdf03ac-784d-4055-870d-f48967a60b89", "firstName": "Xianglong ", "lastName": "Lu", "bio": "Xianglong is a senior deep learning solutions architect at NVIDIA. He has been leading dynamo engineering collaborations in CN region. He specializes in NVIDIA inference software solutions, including Dynamo, Triton, and TensorRT-LLM etc, with deep experience in designing and optimizing large-scale AI inference systems.", "tagLine": "NVIDIA, Key Maintainer of Dynamo ", "profilePicture": "https://cdn.sessionize.com/image/dfe3-400o400o1-XfDUsTetbMM9js7BVaY3g9.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/xianglonglu/", "linkType": "LinkedIn" }, { "title": "Sessionize", "url": "https://sessionize.com/xianglong-lu", "linkType": "Sessionize" } ], "sessions": [ 1224503 ], "fullName": "Xianglong Lu", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "NVIDIA" }, { "questionId": 123918, "answerValue": "Maintainer of Dynamo" } ] }, { "id": "22d235b4-af0f-4f0c-a854-cdccffad169e", "firstName": "Xiao", "lastName": "Zhang", "bio": "Xiao Zhang is the founder of dynamia.ai (focusing on infrastructure, AI, multi-cluster management, cluster lifecycle management (LCM), and the Open Container Initiative (OCI)).\r\nHe is also an active contributor to the community and a cloud native enthusiast. Currently, he is a member of Kubernetes/Kubernetes Special Interest Groups (Kubernetes-sigs) and serves as a maintainer for the Karmada\r\nAt present, he is also a co-author and maintainer of the CNCF HAMi project.\r\nHis GitHub ID is wawa0210.", "tagLine": "dynamia.ai founder, a cloud-native enthusiast and community maintainer, focusing on the AI infrastructure.", "profilePicture": "https://cdn.sessionize.com/image/ba4c-400o400o1-aQkkJnGFb2q1dDCRQJbrP8.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/xiaozhang0210/", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://www.dynamia.ai/", "linkType": "Company_Website" }, { "title": "github", "url": "https://github.com/wawa0210", "linkType": "Other" }, { "title": "Sessionize", "url": "https://sessionize.com/xiao-zhang", "linkType": "Sessionize" } ], "sessions": [ 1221808 ], "fullName": "Xiao Zhang", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "dynamia.ai" }, { "questionId": 123918, "answerValue": "Software Engineer" } ] }, { "id": "a2589860-daa8-43eb-bfa7-19c5c4327034", "firstName": "Xiao", "lastName": "Zhang", "bio": null, "tagLine": null, "profilePicture": null, "isTopSpeaker": false, "links": [], "sessions": [ 1312064 ], "fullName": "Xiao Zhang", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Dynamia.ai" } ] }, { "id": "305b46aa-7ea7-4056-8f3f-8ca103db88c8", "firstName": "Xiaoshuang", "lastName": "Wang", "bio": "I’m an AI infrastructure engineer at Huawei, working on graph pattern development and performance optimization for vLLM-Ascend. By leveraging Ascend-specific operator fusion and graph compilation, I’ve significantly boosted inference efficiency on Ascend. I’m also an active vLLM community contributor, helping co-build the ecosystem.", "tagLine": "Huawei", "profilePicture": "https://cdn.sessionize.com/image/b20c-400o400o1-w6tGhJ2FNkgc8iUVDLM5tj.png", "isTopSpeaker": false, "links": [], "sessions": [ 1222625 ], "fullName": "Xiaoshuang Wang", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Huawei" }, { "questionId": 123918, "answerValue": "Staff Engineer" } ] }, { "id": "6b5d35dd-d75f-4e76-908a-bc21bd7d772e", "firstName": "Xiyuan", "lastName": "Wang", "bio": "10年以上开源社区开发经验，先后参与过众多开源社区与项目，在OpenStack、openEuler、vLLM Ascend社区/项目中担任maintainer职位。目前从事AI开源相关技术的设计和开发工作。", "tagLine": "vLLM-Ascend Committer", "profilePicture": "https://cdn.sessionize.com/image/bb86-400o400o1-fqd2365ycTcPhJ3Kn73PJo.jpg", "isTopSpeaker": false, "links": [ { "title": "Sessionize", "url": "https://sessionize.com/xiyuan-wang", "linkType": "Sessionize" } ], "sessions": [ 1222763 ], "fullName": "Xiyuan Wang", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Huawei" }, { "questionId": 123918, "answerValue": "vLLM Ascend Committer" } ] }, { "id": "05433fa4-5ae1-424b-a513-28d3bb92ae63", "firstName": "Xu", "lastName": "Wang", "bio": "Xu Wang is the vice chair of the Ant Group Open Source Technology Committee and also the head of the Container Infrastructure team. Xu joined Ant Group in 2019, and currently he is responsible for the development of the container-based Cloud-Native infrastructure and the open-source related strategies of Ant Group. \r\nBefore joining Ant Group, Xu was the CTO and co-founder of hyper.sh, a startup on secure container runtime technology, and co-launched Kata Containers project in 2017.", "tagLine": "Ant Group Open Source and Container Infra", "profilePicture": "https://cdn.sessionize.com/image/3853-400o400o1-nJyRtBj51dGBkwpjSU9CVM.jpg", "isTopSpeaker": true, "links": [ { "title": "X (Twitter)", "url": "https://x.com/gnawux", "linkType": "Twitter" }, { "title": "LinkedIn", "url": "https://www.linkedin.com/in/gnawux/", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://www.antgroup.com/en", "linkType": "Company_Website" }, { "title": "Sessionize", "url": "https://sessionize.com/xu-wang", "linkType": "Sessionize" } ], "sessions": [ 1309274 ], "fullName": "Xu Wang", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Ant Group" }, { "questionId": 123918, "answerValue": "Vice Chair, Open Source Technology Committee" }, { "questionId": 138874, "answerValue": "https://events.linuxfoundation.org/wp-content/uploads/2026/08/Ant_Group_logo.png" } ] }, { "id": "82737250-2ba7-402d-87b4-78bb8d7ac2ff", "firstName": "Xuewei", "lastName": "Niu", "bio": "Software developer at Ant Group, Kata Containers committer, open-source enthusiast, focusing on secure containers, kernel specifically in virtualization.", "tagLine": "Empowering Ant Group with Secure Container Solutions", "profilePicture": "https://cdn.sessionize.com/image/6d0b-400o400o1-HzPTgAaKQR4WyvgvjSF7tu.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/justxuewei/", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://www.antgroup.com/en", "linkType": "Company_Website" }, { "title": "Sessionize", "url": "https://sessionize.com/xuewei-niu", "linkType": "Sessionize" } ], "sessions": [ 1220661 ], "fullName": "Xuewei Niu", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "AntGroup" }, { "questionId": 123918, "answerValue": "Software Engineer" } ] }, { "id": "bf6c7299-d682-454a-b453-80e837035988", "firstName": "Yanjun", "lastName": "Chen", "bio": "Yanjun CHEN is an open source expert in China Mobile, and her research mainly focuses on cloud computing, cloud native and telco’s computing force network. She is now one of a technical steering committee members of LF Edge Akraino and actively contributes to open source communities such as CNCF, LF Networking.", "tagLine": "China Mobile", "profilePicture": "https://cdn.sessionize.com/image/bee5-400o400o1-seJABuAgHJp42SFrYpwVeT.jpg", "isTopSpeaker": false, "links": [], "sessions": [ 1224322 ], "fullName": "Yanjun Chen", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "China Mobile Research Institute" }, { "questionId": 123918, "answerValue": "Technical Manager" } ] }, { "id": "db82288a-71ae-4543-83df-49d5fc57227b", "firstName": "Yashi", "lastName": "Su", "bio": "Yashi Su is a software engineer at Alibaba Cloud, focusing on Kubernetes Container Storage Interface (CSI) for object storage. She maintains OSSFS (a FUSE daemon for Alibaba Object Storage Service) used in Cloud-Native scenarios and researches how to improve the read-write performance of object storage for AI workloads. Yashi is also involved in disaster recovery for applications in Kubernetes cluster and the practical application of hardware technologies in Cloud-Native Scenarios.", "tagLine": "Alibaba Cloud, Software Engineer", "profilePicture": "https://cdn.sessionize.com/image/3ec3-400o400o1-pL9mtzpKxWzbwkUAZJYMix.png", "isTopSpeaker": false, "links": [ { "title": "Sessionize", "url": "https://sessionize.com/yashi-su", "linkType": "Sessionize" } ], "sessions": [ 1227863, 1219807 ], "fullName": "Yashi Su", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Alibaba Cloud" }, { "questionId": 123918, "answerValue": "Software Engineer" } ] }, { "id": "8fee27cf-0805-4b12-91d5-350935a3c6ac", "firstName": "Yaya", "lastName": "Xia", "bio": "Xiaoya Xia is a member of the Ant Group OSPO, where she focuses on catalyzing open source success through data-driven insights. Before joining Ant Group, Xiaoya was a PhD at East China Normal University (ECNU), where she concentrated on research into open source ecosystem sustainability. ", "tagLine": "Open Source Analyst, Ant Group ", "profilePicture": "https://cdn.sessionize.com/image/e746-400o400o1-irRRfX45oHJUwig85uBoD.jpg", "isTopSpeaker": true, "links": [ { "title": "X (Twitter)", "url": "https://x.com/YayaXiaoya", "linkType": "Twitter" }, { "title": "LinkedIn", "url": "https://www.linkedin.com/in/xiaoya-xia/", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://www.antgroup.com/en", "linkType": "Company_Website" }, { "title": "Sessionize", "url": "https://sessionize.com/xiaoya-xia", "linkType": "Sessionize" } ], "sessions": [ 1308539, 1309274 ], "fullName": "Yaya Xia", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Ant Group" }, { "questionId": 123918, "answerValue": "Open Source Analyst" }, { "questionId": 138874, "answerValue": "https://events.linuxfoundation.org/wp-content/uploads/2026/08/Ant_Group_logo.png" } ] }, { "id": "a5f0abdd-1eda-4470-97d7-75c7489ca5ac", "firstName": "Yihang", "lastName": "Li", "bio": "Senior Software Engineer at OceanBase", "tagLine": "Senior Software Engineer @ OceanBase", "profilePicture": "https://cdn.sessionize.com/image/a3a3-400o400o1-7W3tJXs1voj9MtMuo4ZLrK.png", "isTopSpeaker": false, "links": [], "sessions": [ 1219464 ], "fullName": "Yihang Li", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "OceanBase, Ant Group" }, { "questionId": 123918, "answerValue": "Software Engineer" } ] }, { "id": "e578e9c1-d7b3-4071-bc8e-a625abaa2b45", "firstName": "Yiheng", "lastName": "Ci", "bio": "I am a passionate professional focusing on Cloud Native technologies, Kubernetes, CI/CD, and DevOps. While always learning, I strive to contribute to efficient and scalable solutions that empower teams and organizations.", "tagLine": "Architect in Tiger Brokers", "profilePicture": "https://cdn.sessionize.com/image/5393-400o400o1-fpw4kfqBMQui81nyEFAhq3.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/yiheng-ci-devops/", "linkType": "LinkedIn" } ], "sessions": [ 1222021 ], "fullName": "Yiheng Ci", "categoryItems": [], "questionAnswers": [ { "questionId": 134288, "answerValue": "Maintainer" } ] }, { "id": "f326d9b7-e8df-4757-8923-58bf9bb6137c", "firstName": "Yin", "lastName": "Ding", "bio": "Yin Ding, an Engineering Leader at VMware, lead the Photon team and brings over 15 years of expertise in large-scale and distributed computing. \r\nAs a co-founder of the CNCF KubeEdge open-source project and the TSC Chair of LF Edge Akraino, Yin Ding has made significant contributions to the advancement of these platforms. He has also contributed to various other open-source projects, including Kubernetes, OpenStack, and Cloud Foundry.", "tagLine": "VMware by Broadcom", "profilePicture": "https://cdn.sessionize.com/image/b229-400o400o1-SsWU36VpfnJufSXgt32aMy.png", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/yin-ding/", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://www.vmware.com/", "linkType": "Company_Website" } ], "sessions": [ 1225125 ], "fullName": "Yin Ding", "categoryItems": [], "questionAnswers": [] }, { "id": "c3488095-867e-43ca-85d8-704cb71ace2f", "firstName": "Yu", "lastName": "Guangye", "bio": "An AI framework engineer at Intel, dedicated to supporting Intel GPUs in PyTorch and advancing the generalization and improvement of PyTorch and its ecosystem. With deep experience in SYCL, XPU backend integration, and performance optimization across heterogeneous platforms, I’ve contributed to enhancing Intel GPU support in PyTorch and collaborated across teams to drive a better developer experience.", "tagLine": "Intel, AI Framework engineer", "profilePicture": "https://cdn.sessionize.com/image/f890-400o400o1-hF7QrXtLv31J5kyLNiw8Lr.jpg", "isTopSpeaker": false, "links": [ { "title": "Company Website", "url": "https://www.intel.com/content/www/us/en/homepage.html", "linkType": "Company_Website" }, { "title": "Sessionize", "url": "https://sessionize.com/yu-guangye", "linkType": "Sessionize" } ], "sessions": [ 1222331 ], "fullName": "Yu Guangye", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Intel" }, { "questionId": 123918, "answerValue": "AI Framework engineer" } ] }, { "id": "23394802-299a-42d2-9aa2-4eb5afe2894d", "firstName": "Yue", "lastName": "Bao", "bio": "Yue Bao serves as a software engineer of Huawei Cloud. She is now working 100% on open source, focusing on lightweight edge for KubeEdge. She is the maintainer of KubeEgde and also the tech leader of KubeEdge SIG Release and Node. Before that, Yue worked on Huawei Cloud Intelligent EdgeFabric Service and participated in multiple edge engineering projects. Yue graduated from Chu Kochen Honors College of Zhejiang University, majoring in information science and electronic engineering.", "tagLine": "KubeCon Europe 2022 Keynote speaker. KubeCon Japan 2025 KeyNote speaker. KubeCon China 2023, 2025 && 2025 session speaker. KubeCon Europe 2025 session speaker", "profilePicture": "https://cdn.sessionize.com/image/40c6-400o400o1-X8M1y36cAf61QvphQFSKmK.jpg", "isTopSpeaker": false, "links": [], "sessions": [ 1225125 ], "fullName": "Yue Bao", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Huawei" }, { "questionId": 123918, "answerValue": "Software Engineer" } ] }, { "id": "04ec9d62-8714-419d-90a2-2fd6cea2045a", "firstName": "Yufeng", "lastName": "Lyu", "bio": null, "tagLine": null, "profilePicture": null, "isTopSpeaker": false, "links": [], "sessions": [ 1224095 ], "fullName": "Yufeng Lyu", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "BAAI" }, { "questionId": 123918, "answerValue": "Researcher of AI Framework" } ] }, { "id": "aac99729-6379-4968-ac02-a13cc6ea9dad", "firstName": "yun", "lastName": "bai", "bio": "2024.11~Now: AI Storage Archetect & System Engineer, Huawei Technologies Co.\r\n2017.7~2024.9: Data Scientist, Intel Products(chengdu) Co, Ltd.", "tagLine": " AI Storage Archetect & System Engineer ", "profilePicture": "https://cdn.sessionize.com/image/a9f6-400o400o1-EtQvU4SDAyww4Np3TExJLd.jpg", "isTopSpeaker": false, "links": [], "sessions": [ 1223545 ], "fullName": "yun bai", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Huawei" }, { "questionId": 123918, "answerValue": "AI Storage Archetect & System Engineer" } ] }, { "id": "3f378b4a-b718-4c9f-ab48-07c8fc1df1d1", "firstName": "Yun", "lastName": "Zhao", "bio": "2025-Now Senior Engineer, Huawei Co,. Ltd\r\n2021-2025 PhD student, Tsinghua University", "tagLine": "Senior Engineer, Huawei Co,. Ltd", "profilePicture": "https://cdn.sessionize.com/image/6969-400o400o1-RQL5ce3dLa54m9jjFp878J.jpg", "isTopSpeaker": false, "links": [ { "title": "Sessionize", "url": "https://sessionize.com/yun-zhao", "linkType": "Sessionize" } ], "sessions": [ 1185605 ], "fullName": "Yun Zhao", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Huawei" }, { "questionId": 123918, "answerValue": "Senior Engineer" } ] }, { "id": "d455f371-8318-45d0-a3be-c10b0af4bee1", "firstName": "Zan", "lastName": "Huang", "bio": "Zan is a developer technology engineer from Nvidia, working on collaborative projects to help customers accelerating recommender systems.", "tagLine": "Nvidia, DevTech Engineer", "profilePicture": "https://cdn.sessionize.com/image/8683-400o400o1-XBmJraie7tMRCXiDtT1k4T.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/pmixer/", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://www.nvidia.com/en-us/", "linkType": "Company_Website" } ], "sessions": [ 1210254 ], "fullName": "Zan Huang", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "NVIDIA" }, { "questionId": 123918, "answerValue": "DevTech Engineer" } ] }, { "id": "c20c52c1-5413-485d-acc8-de4412276233", "firstName": "Zesheng", "lastName": "Zong", "bio": "Zesheng Zong is an AI framework engineer at Huawei. As a PyTorch Ambassador and Co-chair of the PyTorch TAC Accelerator Integration WG, he focuses on accelerator integration and optimization for the PyTorch ecosystem. He promotes global community collaboration, advances framework adaptation for AI accelerators, and helps grow the PyTorch developer community.", "tagLine": "PyTorch Ambassador | PyTorch TAC Accelerator Integration WG Co-chair | AI Framework Engineer", "profilePicture": "https://cdn.sessionize.com/image/e1a2-400o400o1-Qb5Pow2KeEStmeDDUAEXGm.jpg", "isTopSpeaker": true, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/zeshengzong/", "linkType": "LinkedIn" } ], "sessions": [ 1309268, 1309304 ], "fullName": "Zesheng Zong", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Huawei" }, { "questionId": 123918, "answerValue": "AI Framework Engineer" }, { "questionId": 138874, "answerValue": "https://events.linuxfoundation.org/wp-content/uploads/2020/02/Huawei.svg" } ] }, { "id": "a6e8b69b-7ea1-4686-a849-52d913553d69", "firstName": "Zhang", "lastName": "Zhen", "bio": "Zhen Zhang has been working on the cluster management of software applications. he is driving the new cloud native innovation in Alibaba and focus mainly on the application management domain. He is one of main maintainer in OpenKruise project. ", "tagLine": "Alibaba Cloud, staff engineer", "profilePicture": "https://cdn.sessionize.com/image/eb59-400o400o1-hSsq4TySQMuWHP8LJ7mPLY.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/zhenzhangzz/", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://www.alibabacloud.com/", "linkType": "Company_Website" } ], "sessions": [ 1319491, 1224605 ], "fullName": "Zhang Zhen", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Alibaba Cloud" }, { "questionId": 123918, "answerValue": "Staff Engineer" } ] }, { "id": "14ebc104-3e6c-4e50-a410-5398b6d83908", "firstName": "Zhencheng", "lastName": "Lee", "bio": "Zhencheng Lee, Senior Engineer at Huawei Cloud, specializes in Kubernetes, service mesh, and workload in AI. As a member of Huawei Cloud's Cloud Native team, I focus on developing cloud-native open source technologies and building the open source ecosystem. I am the primary developer and maintainer of the CNCF sandbox project Kmesh and Volcano. Actively contribute to several other CNCF projects, with a particular emphasis on service mesh and observability.", "tagLine": "Huawei Cloud Senior R&D engineers", "profilePicture": "https://cdn.sessionize.com/image/4376-400o400o1-M2vjUfSJX5Rr5XEEHaF3SV.jpg", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/%E6%8C%AF%E8%AF%9A-%E6%9D%8E-a23bb533a/", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://www.huawei.com", "linkType": "Company_Website" }, { "title": "Sessionize", "url": "https://sessionize.com/zhencheng-lee", "linkType": "Sessionize" } ], "sessions": [ 1304033 ], "fullName": "Zhencheng Lee", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Huawei Technologies Co., Ltd" }, { "questionId": 123918, "answerValue": "Senior Cloud-Native R&D Engineer" } ] }, { "id": "d3d7d2b1-05c9-4494-be6f-9a2d13eef4e8", "firstName": "Zhuang", "lastName": "Zhang", "bio": "Zhuang Zhang is a senior software engineer for Huawei Cloud. He works 100% on open source, focusing on multi-cluster scheduling, AI application deployment, and CI/CD to accelerate the growth and implementation of cloud-native projects. He is now a maintainer of Karmada, a CNCF incubation project.\r\nHe actively shares his expertise, having presented Karmada’s use in AI/big data workload deployment at KCD 2025, and the integration of Argo with Karmada at KubeCon EU 2026.", "tagLine": "Senior Software engineer, Huawei Cloud", "profilePicture": "https://cdn.sessionize.com/image/5699-400o400o1-U5s2XH8j8vnPxtvQziXaAW.jpg", "isTopSpeaker": false, "links": [ { "title": "Sessionize", "url": "https://sessionize.com/zhuang-zhang", "linkType": "Sessionize" } ], "sessions": [ 1225153 ], "fullName": "Zhuang Zhang", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Huawei Technologies Co. Ltd" }, { "questionId": 123918, "answerValue": "Karmada Maintainer" } ] }, { "id": "ae4766ae-8a22-4193-bb37-a9bc23fa67f0", "firstName": "Zongqing", "lastName": "Li", "bio": "Senior Cloud-Native R&D Engineer at Trip.com, focusing on HPA, VPA, and multi-cluster management in Kubernetes.", "tagLine": "Senior Cloud-Native R&D Engineer @ Trip.com", "profilePicture": "https://cdn.sessionize.com/image/3c1b-400o400o1-HYzVMgyMSFC1icFKiwUd5c.png", "isTopSpeaker": false, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/zongqingli", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://trip.com", "linkType": "Company_Website" } ], "sessions": [ 1222012 ], "fullName": "Zongqing Li", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Trip.com" }, { "questionId": 123918, "answerValue": "Senior Cloud-Native R&D Engineer" } ] }, { "id": "4d2a68f8-5859-49d1-9caa-e488d11baa9b", "firstName": "Zvonko", "lastName": "Kaiser", "bio": "Zvonko Kaiser is a Principal Systems Software Engineer at NVIDIA, specializing in cloud-native confidential computing and GPU security. He serves on the Kata Containers Architecture Committee and the Confidential Containers Technical Steering Committee, and chairs the Kubernetes WG AI Integration. As the primary architect driving both projects at NVIDIA, his work hardens the cloud-native stack from silicon to service, guided by his principle: \"If it's shared, it's vulnerable.\"", "tagLine": "NVIDIA, Principal Systems Software Engineer", "profilePicture": "https://cdn.sessionize.com/image/3a66-400o400o1-Ug7BrLqv3oUC6rhnSaqn6S.png", "isTopSpeaker": true, "links": [ { "title": "LinkedIn", "url": "https://www.linkedin.com/in/zvonkok/", "linkType": "LinkedIn" }, { "title": "Company Website", "url": "https://www.nvidia.com", "linkType": "Company_Website" }, { "title": "Sessionize", "url": "https://sessionize.com/zvonkok", "linkType": "Sessionize" } ], "sessions": [ 1304828 ], "fullName": "Zvonko Kaiser", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "NVIDIA" }, { "questionId": 123918, "answerValue": "Principal Systems Software Engineer" }, { "questionId": 138874, "answerValue": "https://events.linuxfoundation.org/wp-content/uploads/2026/07/nvidia-logo-horz.svg" } ] }, { "id": "782a0eb2-41fe-4175-9452-1ed725201aab", "firstName": "夕宁", "lastName": "王", "bio": "Wang Xining, llm-d project maintainer and senior technical expert of Alibaba AI/LLM Inference Platform, focusing on LLM/AI Infra, MaaS/LLM Inference, Kubernetes, etc. Previously worked in the IBM as tech architect focusing on SOA/Cloud and served as the chairman of the Patent Technology Review Committee. He has more than 100 technical patents in Cloud Computing, and AI related fields. Participated in technology sharing at multiple conferences such as KubeCon/IstioCon, ArchSummit, InfoQ, etc.", "tagLine": "Alibaba AI/LLM Inference Platform, Senior Technical Expert", "profilePicture": "https://cdn.sessionize.com/image/7b86-400o400o1-G9hLuaFwZqr71GgQ31J5Jt.jpg", "isTopSpeaker": false, "links": [ { "title": "Sessionize", "url": "https://sessionize.com/xi-ning-wang", "linkType": "Sessionize" } ], "sessions": [ 1218005 ], "fullName": "夕宁 王", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Alibaba" }, { "questionId": 123918, "answerValue": "Senior Technical Expert" } ] }, { "id": "0a3b95c9-939c-4904-b9b5-c2e72573beef", "firstName": "纪飞", "lastName": "王", "bio": "长期参与云原生与 Kubernetes 生态建设，专注于 GPU 资源管理与集群调度方向，是 HAMi 社区 approver，参与核心功能评审、代码审核与技术决策。其开源贡献涵盖调度器、可观测性等方向，关注云原生系统在生产环境中的可扩展性与可靠性。同时是HAMI-DRA 子项目的主要贡献者。", "tagLine": "密瓜智能研发，HAMi 社区 Approver，HAMi-DRA 主要贡献者", "profilePicture": "https://cdn.sessionize.com/image/9f15-400o400o1-S1KuEkBo5AbKY84R6XSdqq.jpg", "isTopSpeaker": false, "links": [], "sessions": [ 1313959, 1215494 ], "fullName": "纪飞 王", "categoryItems": [], "questionAnswers": [ { "questionId": 123917, "answerValue": "Dynamia" }, { "questionId": 123918, "answerValue": "HAMi Approver" } ] } ], "questions": [ { "id": 123908, "question": "CNCF-hosted software", "questionType": "Short_Text", "sort": 92 }, { "id": 123917, "question": "Company", "questionType": "Short_Text", "sort": 54 }, { "id": 123918, "question": "Speaker Title", "questionType": "Short_Text", "sort": 53 }, { "id": 124487, "question": "Sub-project (if applicable)", "questionType": "Short_Text", "sort": 26 }, { "id": 124494, "question": "Sandbox Project - Project Lightning Talk", "questionType": "Short_Text", "sort": 31 }, { "id": 134032, "question": "Presentation Slides", "questionType": "File_upload", "sort": 74 }, { "id": 134033, "question": "GitHub Repository", "questionType": "Web_address", "sort": 75 }, { "id": 134034, "question": "Website", "questionType": "Web_address", "sort": 76 }, { "id": 134035, "question": "Project Demo", "questionType": "Web_address", "sort": 77 }, { "id": 134036, "question": "Documentation", "questionType": "Web_address", "sort": 78 }, { "id": 134288, "question": "Speaker Company Override", "questionType": "Short_Text", "sort": 103 }, { "id": 138874, "question": "Company URL", "questionType": "Web_address", "sort": 104 } ], "categories": [ { "id": 123931, "title": "Language", "items": [ { "id": 447044, "name": "English", "sort": 1 }, { "id": 447045, "name": "Chinese", "sort": 2 } ], "sort": 1, "type": "session" }, { "id": 123901, "title": "All Sessions", "items": [ { "id": 447046, "name": "AI + ML + Agentic AI + Data Systems", "sort": 1 }, { "id": 447047, "name": "AI Infrastructure + Accelerators + Performance Engineering", "sort": 2 }, { "id": 447048, "name": "Platform Engineering + Cloud Native Architecture", "sort": 3 }, { "id": 447049, "name": "Application Development + Developer Experience", "sort": 4 }, { "id": 447050, "name": "Operations + Observability + Reliability", "sort": 5 }, { "id": 447051, "name": "Security + Privacy + Trusted Computing", "sort": 6 }, { "id": 447052, "name": "Networking + Edge + Distributed Systems", "sort": 7 }, { "id": 447053, "name": "Cloud Infrastructure + Virtualization + Storage", "sort": 8 }, { "id": 447054, "name": "Emerging Technologies + Research + Advanced Topics", "sort": 9 }, { "id": 447055, "name": "Community + Open Source + Getting Started", "sort": 10 }, { "id": 447056, "name": "Hardware Enablement + Diversification", "sort": 11 }, { "id": 482870, "name": "Keynote Sessions", "sort": 12 }, { "id": 485106, "name": "Breaks", "sort": 13 }, { "id": 485107, "name": "Registration + Badge Pick-up", "sort": 14 }, { "id": 485108, "name": "Maintainer Track", "sort": 15 }, { "id": 485109, "name": "Project Opportunities", "sort": 16 }, { "id": 485110, "name": "Solutions Showcase", "sort": 17 }, { "id": 485111, "name": "Experiences", "sort": 18 }, { "id": 485112, "name": "⚡Lightning Talks", "sort": 19 }, { "id": 485113, "name": "📚 Tutorials", "sort": 20 }, { "id": 500857, "name": "Sponsor Demos", "sort": 21 }, { "id": 502427, "name": "CNCF-Hosted Co-located Events", "sort": 22 }, { "id": 503598, "name": "Sponsor-hosted Co-located Events", "sort": 23 } ], "sort": 5, "type": "session" }, { "id": 123902, "title": "Level", "items": [ { "id": 446827, "name": "Any", "sort": 1 }, { "id": 446825, "name": "Beginner", "sort": 2 }, { "id": 446826, "name": "Intermediate", "sort": 3 }, { "id": 446824, "name": "Advanced", "sort": 4 } ], "sort": 7, "type": "session" }, { "id": 124483, "title": "Graduated Project - Maintainer Track", "items": [ { "id": 449228, "name": "Argo", "sort": 1 }, { "id": 449245, "name": "cert-manager", "sort": 2 }, { "id": 449246, "name": "Cilium", "sort": 3 }, { "id": 449247, "name": "Cloudevents", "sort": 4 }, { "id": 449248, "name": "Containerd", "sort": 5 }, { "id": 449249, "name": "CoreDNS", "sort": 6 }, { "id": 449250, "name": "cri-o", "sort": 7 }, { "id": 449251, "name": "Crossplane", "sort": 8 }, { "id": 449252, "name": "CubeFS", "sort": 9 }, { "id": 449253, "name": "Dapr", "sort": 10 }, { "id": 449254, "name": "Dragonfly", "sort": 11 }, { "id": 449255, "name": "Envoy", "sort": 12 }, { "id": 449256, "name": "etcd", "sort": 13 }, { "id": 449257, "name": "Falco", "sort": 14 }, { "id": 449258, "name": "Fluentd", "sort": 15 }, { "id": 449259, "name": "Flux", "sort": 16 }, { "id": 449260, "name": "Harbor", "sort": 17 }, { "id": 449229, "name": "Helm", "sort": 18 }, { "id": 449230, "name": "in-toto", "sort": 19 }, { "id": 449231, "name": "Istio", "sort": 20 }, { "id": 449232, "name": "Jaeger", "sort": 21 }, { "id": 449233, "name": "Keda", "sort": 22 }, { "id": 449234, "name": "Knative", "sort": 23 }, { "id": 449235, "name": "KubeEdge", "sort": 24 }, { "id": 449236, "name": "Linkerd", "sort": 25 }, { "id": 449237, "name": "Open Policy Agent (OPA)", "sort": 26 }, { "id": 449238, "name": "Prometheus", "sort": 27 }, { "id": 449239, "name": "Rook", "sort": 28 }, { "id": 449240, "name": "SPIFFE", "sort": 29 }, { "id": 449241, "name": "SPIRE", "sort": 30 }, { "id": 449242, "name": "The Update Framework (TUF)", "sort": 31 }, { "id": 449243, "name": "TiKV", "sort": 32 }, { "id": 449244, "name": "Vitess", "sort": 33 } ], "sort": 22, "type": "session" }, { "id": 124484, "title": "Incubating Project - Maintainer Track", "items": [ { "id": 449293, "name": "ArtifactHub", "sort": 1 }, { "id": 449294, "name": "Backstage", "sort": 2 }, { "id": 449295, "name": "Buildpacks", "sort": 3 }, { "id": 449296, "name": "Chaos Mesh", "sort": 4 }, { "id": 449297, "name": "Cloud Custodian", "sort": 5 }, { "id": 449298, "name": "Container Network Interface", "sort": 6 }, { "id": 449299, "name": "Contour", "sort": 7 }, { "id": 449261, "name": "Cortex", "sort": 8 }, { "id": 449262, "name": "Emissary-ingress", "sort": 9 }, { "id": 470680, "name": "Fluid", "sort": 10 }, { "id": 449263, "name": "gRPC", "sort": 11 }, { "id": 449264, "name": "Karmada", "sort": 12 }, { "id": 449265, "name": "Keycloak", "sort": 13 }, { "id": 449266, "name": "kserve", "sort": 14 }, { "id": 449267, "name": "Kubeflow", "sort": 15 }, { "id": 449268, "name": "Kubescape", "sort": 16 }, { "id": 449269, "name": "KubeVela", "sort": 17 }, { "id": 449270, "name": "KubeVirt", "sort": 18 }, { "id": 449271, "name": "Kyverno", "sort": 19 }, { "id": 449272, "name": "Lima", "sort": 20 }, { "id": 449273, "name": "Litmus", "sort": 21 }, { "id": 449274, "name": "Longhorn", "sort": 22 }, { "id": 449275, "name": "Metal3.io", "sort": 23 }, { "id": 449276, "name": "NATS", "sort": 24 }, { "id": 449316, "name": "Notary", "sort": 25 }, { "id": 449317, "name": "OpenCost", "sort": 26 }, { "id": 449318, "name": "OpenFeature", "sort": 27 }, { "id": 449319, "name": "OpenFGA", "sort": 28 }, { "id": 449320, "name": "OpenKruise", "sort": 29 }, { "id": 449321, "name": "OpenTelemetry", "sort": 30 }, { "id": 449322, "name": "OpenYurt", "sort": 31 }, { "id": 449323, "name": "Operator Framework", "sort": 32 }, { "id": 449324, "name": "Strimzi", "sort": 33 }, { "id": 449325, "name": "Thanos", "sort": 34 }, { "id": 449326, "name": "Volcano", "sort": 35 }, { "id": 449327, "name": "wasmCloud", "sort": 36 } ], "sort": 23, "type": "session" }, { "id": 124485, "title": "Working Group - Maintainer Track", "items": [ { "id": 449328, "name": "AI Conformance", "sort": 1 }, { "id": 449329, "name": "AI Gateway", "sort": 2 }, { "id": 449330, "name": "AI Integration", "sort": 3 }, { "id": 449331, "name": "Batch", "sort": 4 }, { "id": 449277, "name": "Checkpoint Restore", "sort": 5 }, { "id": 449278, "name": "Data Protection", "sort": 6 }, { "id": 449279, "name": "Device Management", "sort": 7 }, { "id": 449280, "name": "etcd Operator", "sort": 8 }, { "id": 449281, "name": "LTS", "sort": 9 }, { "id": 449282, "name": "Node Lifecycle", "sort": 10 }, { "id": 449283, "name": "Serving", "sort": 11 }, { "id": 449284, "name": "Structured Logging", "sort": 12 } ], "sort": 24, "type": "session" }, { "id": 124486, "title": "Special Interest Group - Maintainer Track", "items": [ { "id": 449285, "name": "API Machinery", "sort": 1 }, { "id": 449286, "name": "Apps", "sort": 2 }, { "id": 449287, "name": "Architecture", "sort": 3 }, { "id": 449288, "name": "Auth", "sort": 4 }, { "id": 449289, "name": "Autoscaling", "sort": 5 }, { "id": 449290, "name": "CLI", "sort": 6 }, { "id": 449291, "name": "Cloud Provider", "sort": 7 }, { "id": 449292, "name": "Cluster Lifecycle", "sort": 8 }, { "id": 449300, "name": "Contributor Experience", "sort": 9 }, { "id": 449301, "name": "Docs", "sort": 10 }, { "id": 449302, "name": "etcd", "sort": 11 }, { "id": 449303, "name": "Instrumentation", "sort": 12 }, { "id": 449304, "name": "K8s Infra", "sort": 13 }, { "id": 449305, "name": "Multicluster", "sort": 14 }, { "id": 449306, "name": "Network", "sort": 15 }, { "id": 449307, "name": "Node", "sort": 16 }, { "id": 449308, "name": "Release", "sort": 17 }, { "id": 449309, "name": "Scalability", "sort": 18 }, { "id": 449310, "name": "Scheduling", "sort": 19 }, { "id": 449311, "name": "Security", "sort": 20 }, { "id": 449312, "name": "Storage", "sort": 21 }, { "id": 449313, "name": "Testing", "sort": 22 }, { "id": 449314, "name": "UI", "sort": 23 }, { "id": 449315, "name": "Windows", "sort": 24 } ], "sort": 25, "type": "session" }, { "id": 124488, "title": "Technical Advisory Group - Maintainer Track", "items": [ { "id": 449332, "name": "Developer Experience", "sort": 1 }, { "id": 449333, "name": "Infrastructure", "sort": 2 }, { "id": 449334, "name": "Operational Resilience", "sort": 3 }, { "id": 449335, "name": "Security & Compliance", "sort": 4 }, { "id": 449336, "name": "Workloads Foundation", "sort": 5 } ], "sort": 27, "type": "session" }, { "id": 124490, "title": "Graduated Project - Project Lightning Talk", "items": [ { "id": 449344, "name": "Argo", "sort": 1 }, { "id": 449377, "name": "cert-manager", "sort": 2 }, { "id": 449378, "name": "Cilium", "sort": 3 }, { "id": 449379, "name": "Cloudevents", "sort": 4 }, { "id": 449380, "name": "Containerd", "sort": 5 }, { "id": 449381, "name": "CoreDNS", "sort": 6 }, { "id": 449382, "name": "cri-o", "sort": 7 }, { "id": 449383, "name": "Crossplane", "sort": 8 }, { "id": 449384, "name": "CubeFS", "sort": 9 }, { "id": 449385, "name": "Dapr", "sort": 10 }, { "id": 449386, "name": "Dragonfly", "sort": 11 }, { "id": 449387, "name": "Envoy", "sort": 12 }, { "id": 449388, "name": "etcd", "sort": 13 }, { "id": 449389, "name": "Falco", "sort": 14 }, { "id": 449390, "name": "Fluentd", "sort": 15 }, { "id": 449391, "name": "Flux", "sort": 16 }, { "id": 449392, "name": "Harbor", "sort": 17 }, { "id": 449396, "name": "Helm", "sort": 18 }, { "id": 449397, "name": "in-toto", "sort": 19 }, { "id": 449398, "name": "Istio", "sort": 20 }, { "id": 449399, "name": "Jaeger", "sort": 21 }, { "id": 449400, "name": "Keda", "sort": 22 }, { "id": 449401, "name": "Knative", "sort": 23 }, { "id": 449402, "name": "KubeEdge", "sort": 24 }, { "id": 449595, "name": "Kyverno", "sort": 25 }, { "id": 449403, "name": "Linkerd", "sort": 26 }, { "id": 449404, "name": "Open Policy Agent (OPA)", "sort": 27 }, { "id": 449405, "name": "Prometheus", "sort": 28 }, { "id": 449406, "name": "Rook", "sort": 29 }, { "id": 449407, "name": "SPIFFE", "sort": 30 }, { "id": 449408, "name": "SPIRE", "sort": 31 }, { "id": 449409, "name": "The Update Framework (TUF)", "sort": 32 }, { "id": 449410, "name": "TiKV", "sort": 33 }, { "id": 449411, "name": "Vitess", "sort": 34 } ], "sort": 29, "type": "session" }, { "id": 124491, "title": "Incubating Project - Project Lightning Talk", "items": [ { "id": 449345, "name": "ArtifactHub", "sort": 1 }, { "id": 449346, "name": "Backstage", "sort": 2 }, { "id": 449347, "name": "Buildpacks", "sort": 3 }, { "id": 449348, "name": "Chaos Mesh", "sort": 4 }, { "id": 449349, "name": "Cloud Custodian", "sort": 5 }, { "id": 449350, "name": "Container Network Interface", "sort": 6 }, { "id": 449351, "name": "Contour", "sort": 7 }, { "id": 449352, "name": "Cortex", "sort": 8 }, { "id": 449353, "name": "Emissary-ingress", "sort": 9 }, { "id": 449354, "name": "gRPC", "sort": 10 }, { "id": 449355, "name": "Karmada", "sort": 11 }, { "id": 449356, "name": "Keycloak", "sort": 12 }, { "id": 449357, "name": "kserve", "sort": 13 }, { "id": 449358, "name": "Kubeflow", "sort": 14 }, { "id": 449359, "name": "Kubescape", "sort": 15 }, { "id": 449360, "name": "KubeVela", "sort": 16 }, { "id": 449361, "name": "KubeVirt", "sort": 17 }, { "id": 449363, "name": "Lima", "sort": 18 }, { "id": 449364, "name": "Litmus", "sort": 19 }, { "id": 449365, "name": "Longhorn", "sort": 20 }, { "id": 449366, "name": "Metal3.io", "sort": 21 }, { "id": 449367, "name": "NATS", "sort": 22 }, { "id": 449368, "name": "Notary", "sort": 23 }, { "id": 449369, "name": "OpenCost", "sort": 24 }, { "id": 449370, "name": "OpenFeature", "sort": 25 }, { "id": 449371, "name": "OpenFGA", "sort": 26 }, { "id": 449372, "name": "OpenKruise", "sort": 27 }, { "id": 449373, "name": "OpenTelemetry", "sort": 28 }, { "id": 449374, "name": "OpenYurt", "sort": 29 }, { "id": 449375, "name": "Operator Framework", "sort": 30 }, { "id": 449376, "name": "Strimzi", "sort": 31 }, { "id": 449393, "name": "Thanos", "sort": 32 }, { "id": 449394, "name": "Volcano", "sort": 33 }, { "id": 449395, "name": "wasmCloud", "sort": 34 } ], "sort": 30, "type": "session" }, { "id": 124492, "title": "Special Interest Group - Project Lightning Talk", "items": [ { "id": 449412, "name": "API Machinery", "sort": 1 }, { "id": 449413, "name": "Apps", "sort": 2 }, { "id": 449414, "name": "Architecture", "sort": 3 }, { "id": 449415, "name": "Auth", "sort": 4 }, { "id": 449416, "name": "Autoscaling", "sort": 5 }, { "id": 449417, "name": "CLI", "sort": 6 }, { "id": 449418, "name": "Cloud Provider", "sort": 7 }, { "id": 449419, "name": "Cluster Lifecycle", "sort": 8 }, { "id": 449420, "name": "Contributor Experience", "sort": 9 }, { "id": 449421, "name": "Docs", "sort": 10 }, { "id": 449422, "name": "etcd", "sort": 11 }, { "id": 449423, "name": "Instrumentation", "sort": 12 }, { "id": 449424, "name": "K8s Infra", "sort": 13 }, { "id": 449425, "name": "Multicluster", "sort": 14 }, { "id": 449426, "name": "Network", "sort": 15 }, { "id": 449427, "name": "Node", "sort": 16 }, { "id": 449428, "name": "Release", "sort": 17 }, { "id": 449429, "name": "Scalability", "sort": 18 }, { "id": 449430, "name": "Scheduling", "sort": 19 }, { "id": 449431, "name": "Security", "sort": 20 }, { "id": 449432, "name": "Storage", "sort": 21 }, { "id": 449433, "name": "Testing", "sort": 22 }, { "id": 449434, "name": "UI", "sort": 23 }, { "id": 449435, "name": "Windows", "sort": 24 } ], "sort": 32, "type": "session" }, { "id": 124493, "title": "Technical Advisory Group - Project Lightning Talk", "items": [ { "id": 449436, "name": "Developer Experience", "sort": 1 }, { "id": 449437, "name": "Infrastructure", "sort": 2 }, { "id": 449438, "name": "Operational Resilience", "sort": 3 }, { "id": 449439, "name": "Security & Compliance", "sort": 4 }, { "id": 449440, "name": "Workloads Foundation", "sort": 5 } ], "sort": 33, "type": "session" }, { "id": 134029, "title": "Lightning Talk Track", "items": [ { "id": 485119, "name": "AI + ML + Agentic AI + Data Systems", "sort": 1 }, { "id": 485123, "name": "AI Infrastructure + Accelerators + Performance Engineering", "sort": 2 }, { "id": 485120, "name": "Platform Engineering + Cloud Native Architecture", "sort": 3 }, { "id": 485114, "name": "Application Development + Developer Experience", "sort": 4 }, { "id": 485115, "name": "Operations + Observability + Reliability", "sort": 5 }, { "id": 485121, "name": "Security + Privacy + Trusted Computing", "sort": 6 }, { "id": 485116, "name": "Networking + Edge + Distributed Systems", "sort": 7 }, { "id": 485124, "name": "Cloud Infrastructure + Virtualization + Storage", "sort": 8 }, { "id": 485117, "name": "Emerging Technologies + Research + Advanced Topics", "sort": 9 }, { "id": 485122, "name": "Community + Open Source + Getting Started", "sort": 10 }, { "id": 485118, "name": "Hardware Enablement + Diversification", "sort": 11 } ], "sort": 93, "type": "session" }, { "id": 134030, "title": "Tutorial Track", "items": [ { "id": 485130, "name": "AI + ML + Agentic AI + Data Systems", "sort": 1 }, { "id": 485134, "name": "AI Infrastructure + Accelerators + Performance Engineering", "sort": 2 }, { "id": 485131, "name": "Platform Engineering + Cloud Native Architecture", "sort": 3 }, { "id": 485125, "name": "Application Development + Developer Experience", "sort": 4 }, { "id": 485126, "name": "Operations + Observability + Reliability", "sort": 5 }, { "id": 485132, "name": "Security + Privacy + Trusted Computing", "sort": 6 }, { "id": 485135, "name": "Networking + Edge + Distributed Systems", "sort": 7 }, { "id": 485127, "name": "Cloud Infrastructure + Virtualization + Storage", "sort": 8 }, { "id": 485133, "name": "Emerging Technologies + Research + Advanced Topics", "sort": 9 }, { "id": 485128, "name": "Community + Open Source + Getting Started", "sort": 10 }, { "id": 485129, "name": "Hardware Enablement + Diversification", "sort": 11 } ], "sort": 94, "type": "session" }, { "id": 134286, "title": "Project Benefit Designation", "items": [ { "id": 486111, "name": "KubeEdge", "sort": 1 }, { "id": 486112, "name": "Litmus", "sort": 2 }, { "id": 486116, "name": "Karmada", "sort": 3 }, { "id": 486113, "name": "OpenEverest", "sort": 4 }, { "id": 486106, "name": "Dragonfly", "sort": 5 }, { "id": 486107, "name": "Dapr", "sort": 6 }, { "id": 486114, "name": "Argo", "sort": 7 }, { "id": 486108, "name": "Cilium", "sort": 8 }, { "id": 486117, "name": "Volcano", "sort": 9 }, { "id": 486109, "name": "OpenKruise", "sort": 10 }, { "id": 486115, "name": "Apps", "sort": 11 }, { "id": 486110, "name": "Infrastructure", "sort": 12 }, { "id": 486118, "name": "Atlantis", "sort": 13 }, { "id": 486119, "name": "HAMi", "sort": 14 }, { "id": 506816, "name": "Perses", "sort": 15 }, { "id": 507394, "name": "Headlamp", "sort": 16 } ], "sort": 95, "type": "session" }, { "id": 140057, "title": "Sponsored", "items": [ { "id": 508425, "name": "Sponsored", "sort": 1 } ], "sort": 97, "type": "session" } ], "rooms": [ { "id": 80134, "name": "1F | Mandarin Hall I", "sort": 0 }, { "id": 80135, "name": "1F | Mandarin Hall II", "sort": 1 }, { "id": 80136, "name": "7F | Grand Ballroom II + III", "sort": 2 }, { "id": 80137, "name": "7F | Pearl Hall", "sort": 3 }, { "id": 85688, "name": "5F | 5B + C", "sort": 4 }, { "id": 85877, "name": "1F Foyer", "sort": 5 }, { "id": 85878, "name": "7F | Grand Ballroom I", "sort": 6 }, { "id": 87761, "name": "5F | 5D + E", "sort": 7 }, { "id": 88309, "name": "7F Foyer", "sort": 8 }, { "id": 88631, "name": "5F | Jiu Zhou Hall", "sort": 9 }, { "id": 88856, "name": "Off-Site Location", "sort": 10 } ] }