Samaresh Kumar Singh
Principal Engineer, HP – Distributed Systems, Edge AI/ML, Cloud-Native Orchestration and Security
Actions
Samaresh Kumar Singh is an Engineering Principal at HP Inc. with more than 21 years of experience designing and implementing large-scale distributed systems, cloud-native platform architectures, and secure Edge AI / ML systems. His expertise spans agentic AI systems, GenAI / LLMs, Edge AI, federated and privacy-preserving learning, and secure hybrid cloud–edge computing.
In addition to building intelligent systems, Samaresh specializes in cybersecurity, Zero Trust architectures, and AI governance frameworks for production environments. He has led initiatives focused on securing AI pipelines across the full lifecycle from data ingestion and model training to deployment, monitoring, and continuous validation. His work integrates secure MLOps practices, model risk management, AI supply chain security, identity-centric access control, and runtime protection for distributed AI systems operating at the edge.
Throughout his career, he has architected high-throughput, low-latency services and resilient orchestration frameworks that deliver intelligence closer to devices and data sources while maintaining strong security, compliance, and governance controls. He actively contributes to the technical community, serving as a book review editor for Apress and Manning, and has reviewed more than 50 research papers for IEEE conferences and journals spanning AI/ML, distributed systems, edge computing, cybersecurity, and AI risk management.
Samaresh is passionate about translating academic research into secure, production-grade systems and mentoring engineers and researchers in building trustworthy, scalable, and governance-aligned AI platforms.
Running Kubernetes in Production at the Edge: Lightweight Clusters &Real-World Challenges
Edge environments are messy—small devices, unreliable networks, and no on-site engineers when something breaks. Yet companies like Chick-fil-A, retail chains, and manufacturing plants are proving that Kubernetes can run reliably at the edge if designed correctly.
In this talk, we’ll explore how to run Kubernetes in production on resource-constrained edge nodes using lightweight distros like K3s and cloud-connected frameworks like KubeEdge. We’ll cover real-world challenges (limited CPU/RAM, intermittent connectivity, security, and operational scale), the architectural patterns that work, and best practices for keeping edge clusters resilient and autonomous.
The session includes a live demo: deploying a small Golang microservice to a K3s edge node on a Raspberry Pi, pushing images to a local edge registry, and demonstrating Kubernetes self-healing in a simulated failure scenario.
Attendees will leave with a practical blueprint for building production-ready edge platforms using Kubernetes—and a clear understanding of when and why edge Kubernetes makes sense.
Kubernetes at the Edge: Running Production Workloads with K3s and KubeEdge
As AI/ML and data-driven applications increasingly move closer to users and sensors, edge computing has become a critical architectural shift. In this lightning talk, Samaresh Kumar Singh shares real-world insights from running Kubernetes in production at the edge using K3s and KubeEdge. Drawing on his experience building Hydra - a distributed edge orchestration platform—he covers the core architectural patterns, production challenges (resource constraints, connectivity, security), and best practices for deploying containerized Golang services at the edge. A live demo will show how lightweight Kubernetes setups enable scalable, resilient edge deployments that stay functional even in cloud-disconnected scenarios. This session is ideal for developers, architects, and DevOps engineers looking to extend Kubernetes beyond the datacenter.
From Sensors to Edge AI: Building Real-Time IoT Intelligence at Scale
Modern IoT systems are no longer just “devices sending data to the cloud.” To deliver real-time insights, anomaly detection, and closed-loop control, we increasingly need intelligence at the edge—right where data is generated.
In this session, I’ll walk through how to design and build an IoT architecture that pushes analytics and AI/ML to the edge while still integrating cleanly with cloud services. Drawing on production experience from industrial IoT, smart buildings, and edge AI platforms, we’ll cover:
How to structure an IoT stack from devices → edge gateways → cloud
When and why to move analytics and ML inference to the edge
Practical patterns for messaging (MQTT, pub/sub), streaming, and command/control
Designing for reliability: offline operation, reconnection, and buffering
Security basics for IoT (device identity, certificates, secure update paths)
Real examples of using edge intelligence for predictive maintenance and anomaly detection
The talk will be vendor-neutral and code-light, focusing on architecture, patterns, and trade-offs that software engineers and architects can take back to their own IoT or edge projects.
esigning High-Performance Distributed AI Systems: Cloud-Native Architecture Patterns
Modern AI applications need more than strong models—they need reliable, low-latency, cloud-native systems capable of running at global scale. In this session, Samaresh shares engineering and architectural lessons from building Hydra a next-generation distributed AI/ML orchestration platform that runs across hybrid cloud and edge environments.
The talk breaks down practical patterns for designing robust AI microservices:
1. event-driven and message-driven architectures,
2. containerized ML pipelines,
3. GPU/CPU/NPU workload scheduling,
4. real-time telemetry and observability,
5. concurrency, resilience, and back-pressure strategies,
6. achieving low-latency inference in unreliable network conditions.
Attendees will also learn how AI workloads benefit from cloud-native principles—Kubernetes, service meshes, asynchronous messaging, distributed caching, secure communication, and zero-downtime deployments.
Building Real-Time Edge AI Systems: From Cloud-Native to On-Device Intelligence
Modern applications increasingly demand real-time intelligence, low latency, and privacy-aware analytics—capabilities that centralized cloud systems alone cannot deliver. This session walks through how to design and build production-grade Edge AI systems using cloud-native principles, container orchestration, and lightweight ML runtimes running on CPUs/GPUs/NPUs.
Drawing from hands-on experience building HP Hydra, HP AI Studio, and HP Z Boost, the talk covers:
How cloud–edge architectures differ from traditional cloud apps
Deploying AI/ML workloads across heterogeneous hardware
Real-time inference, data locality, and intelligent task routing
Observability, CI/CD, and DevOps considerations for edge fleets
Security foundations: PKI, TLS/mTLS, attestation, and zero-trust edge
Lessons learned scaling AI at the edge for enterprise customers
This session is designed to be useful to developers, architects, students, and anyone curious about the future of distributed AI.
Deploying AI at the Edge with Kubernetes: Real-Time ML and Scaling Challenges
Running AI/ML workloads at the edge introduces challenges that traditional cloud-only architectures cannot solve latency, unreliable networking, heterogeneous hardware, and distributed coordination. This session shares practical lessons from building High performance system, a cloud-native, Kubernetes-aligned edge orchestration framework designed to run AI/ML inference and distributed compute across thousands of GPU/NPU/CPU edge nodes.
The talk covers:
1. Kubernetes patterns for deploying and managing edge AI workloads
2 Running ML pipelines across mixed hardware with containerized microservices
3. Real-time inference, scheduling, observability, and reliability at the edge
4. How we combined cloud-native DevOps + AI/ML + distributed systems to achieve resilient deployment
5. Practical failures, debugging war stories, and performance optimizations
6. Attendees will take away actionable patterns for building high-performance, production-ready edge AI systems using Kubernetes and CNCF tools.
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