Tessa Pham
Senior Software Engineer at Bloomberg
Actions
Tessa Pham is a Senior Software Engineer at Bloomberg, where she works on cloud-native training infrastructure for the company’s Data Science Platform and has previously developed its inference platform. Her work supports scalable, reliable ML model training, deployment and serving across the organization. Tessa is an open source contributor to KServe and Karmada. With a background in linguistics, she’s particularly interested in cloud-native ML systems that support LLMs and NLP pipelines.
Multi-cluster Orchestration System: Karmada Updates and Use Cases
Karmada (Kubernetes Armada) is a Kubernetes management system that enables you to run your cloud-native applications across multiple Kubernetes clusters and clouds.
In this presentation, the maintainer of the Karmada project will share:
- A Brief introduction to Karmada.
- New features over the last year
- Application Priority Scheduling
- Federated ResourceQuota Enforcement
- Stateful Application Cluster Failover
- AI Jobs Scheduling Enhancements
- Remarkable Performance Optimization
- Karmada Dashboard Release
- Karmada Operator Enhancement
- Real-world case studies
- Overview of the community
- Roadmap
- QA
Serving the Future: KServe’s Next Chapter Hosting LLMs & GenAI Models (with Fun Drawings!)
In the rapidly evolving generative AI landscape, KServe has emerged as a pivotal platform for deploying and managing LLMs at scale. KServe simplifies deploying ML models on Kubernetes, but there’s so much more to the story than predictor pods and YAML files. With its newly expanded capabilities, KServe is ready to host the next generation of AI workloads, including LLMs and other generative AI applications.
As both maintainers of KServe and daily practitioners running it in Bloomberg’s clusters, we bring firsthand insights into how users utilize KServe to deploy advanced LLM features in production across hybrid environments. This session will delve into KServe's latest features tailored for generative AI. We will offer insights into its enhanced serving runtimes, scalability improvements, and integration strategies. Attendees will gain practical knowledge about deploying and scaling generative models using KServe, informed by real-world experiences and the lessons we’ve learned.
Engaging the KServe Community: The Impact of Integrating Solutions with Standardized CNCF Projects
Building a new solution and contemplating whether or not the OSS path is right for you? Wondering where to get started with a large cloud initiative and where the pitfalls may lie? Curious to know all the benefits waiting if your organization embraces a rich CNCF ecosystem?
In this talk we will discuss the trade-offs between building a product on a full OSS platform vs. a DIY approach. We will delve into the issues of working with internal stakeholders or partners to embrace an OSS community and will cover the benefits and scaling factors that come when embracing open standards.
We will use the recent integration of NVIDIA NIM into KServe as a case study and talk through the trials and tribulations that paid off in a win-win-win situation for our solutions, the OSS projects, and our users. We will cover Kubeflow, Knative, Istio, KServe, and wg-serve as well as a network of companies building enterprise K8s platforms and enterprise AI applications on top of these foundations.
The Hitchhiker's Guide to Kubernetes Platforms: Don’t Panic, Just Launch!
AI development cycles have a vast expanse of specific requirements and needs. Built with the right tools, a cloud-native computing platform on top of Kubernetes unleashes galactic possibilities to develop models quicker and easier than ever before. Join us for a celestial journey (with fun drawings) building an AI platform. At our mission’s end, we’ll discover a robust, scalable, secure platform ensuring stability, observability, and consistent UX via a web portal and self-service APIs.
At Bloomberg, we’ve been modularizing our Data Science Platform solar system with distinctive APIs (multi-cluster deployment, version control, debugging, resource management) that orbit together to efficiently manage and expose information.
We’ll use an ML inference platform to explore the cosmic realms of this implementation, using open source projects as guiding stars. We’ll provide guidance and ideas for other engineers to construct a stellar cloud-native platform and measure its universal success.
CNCF-hosted Co-located Events Europe 2025 Sessionize Event
KubeCon + CloudNativeCon North America 2024 Sessionize Event
CNCF-hosted Co-located Events Europe 2024 Sessionize Event
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top