• Architect and implement HPC clusters for AI, simulation, and distributed training using Kubernetes and schedulers like Slurm. • Integrate NVIDIA Hopper and Blackwell‑class GPUs with NVLink/NVSwitch and InfiniBand/RoCE. • Deploy and manage GPU Operator and Network Operator for large fleets. • Design and validate cloud‑native HPC environments with low latency and high bandwidth. • Define and document reference architectures for AI model training and MLOps. • Collaborate with NVIDIA and other partners to evaluate new GPU generations and software stacks. • Benchmark performance, track down bottlenecks, and recommend concrete changes. • Lead design sessions and architecture reviews with customers focused on performance and reliability.