NVIDIA Corporation seeks a Software Engineer to bring up, benchmark, and optimize distributed AI workloads at scale. You will validate large multi-GPU clusters, triage failures, and tune runtimes across PyTorch, CUDA, NCCL, and Megatron/Nemo stacks.
Collaboration with framework, systems, and platform teams is essential to deliver data-driven performance insights and robust qualification workflows. Candidates have 3+ years in AI or HPC software, solid Python and C/C++, and experience with
📌 Ai Infrastructure Engineer Redmond
🏢 Nvidia
📍 Redmond
Reply to this offer
Impress this employer describing Your skills and abilities, fill out the form below and leave Your personal touch in the presentation letter.