Clement Pakkam Isaac
Solutions Architect | Production LLM Inference on Kubernetes and AKS
📋 Biography
Clement Pakkam Isaac is a Solutions Architect at NVIDIA specializing in enterprise AI. He partners with organizations to accelerate AI adoption through scalable, GPU-accelerated solutions that bridge NVIDIA's advanced tech with enterprise needs.
✨ High-Impact Contributions 2
At Microsoft Build 2026 in San Francisco, I co-presented and facilitated a hands-on lab on building AI-powered video applications with NVIDIA vision agents on Azure Kubernetes Service (AKS). Participants used the NVIDIA Video Search and Summarization blueprint and combined vision-language and large language models to index, search, and summarize video streams through natural-language interaction. I explained the solution and AKS deployment architecture, guided participants through the exercises, helped resolve technical and setup questions, and connected the lab to practical enterprise application patterns. Participants gained reusable, hands-on experience deploying accelerated AI workloads on AKS.
I co-presented a Microsoft Reactor livestream with Jonathan Tong of Microsoft on building production-ready LLM inference with NVIDIA Dynamo and Azure Kubernetes Service. Using an e-commerce recommendation scenario, I explained multi-node orchestration, GPU scheduling, low-latency communication, distributed inference, and cost-efficient scaling. I translated these advanced concepts into practical architecture and deployment guidance that developers and cloud-native practitioners could apply to enterprise AI workloads. This was the Americas-friendly delivery of a two-time-zone program, and the recording remains freely available for continued learning.