NVIDIA role lists core skills
A recent NVIDIA System Software Engineer job listing emphasizes Python, C++, CUDA, PyTorch/TensorFlow and familiarity with agentic frameworks like LangChain—reconfirming that deep coding + GPU programming and ML framework fluency are baseline expectations. The posting suggests labs expect engineers to own both model code and production performance. (x.com)
NVIDIA posted a role titled "System Software Engineer, GPU — New College Graduate 2026" for Pune and Bengaluru with job code JR2015112 and a "Posted Today" timestamp on its careers site. (nvidia.wd5.myworkdayjobs.com) The listing assigns hires to develop autonomous AI agents and to contribute to the next‑generation GPU system software stack and tooling for production‑ready agent deployments. (nvidia.wd5.myworkdayjobs.com) The ad’s "Ways to stand out" section specifically calls for demonstrated experience building or deploying agentic AI systems and prior contributions to open‑source AI ecosystems such as agent frameworks or inference‑optimization projects. (nvidia.wd5.myworkdayjobs.com) Parallel senior NVIDIA listings continue that pattern: recent Senior Deep‑Learning and Inference roles require production experience with PyTorch/TensorFlow and emphasize performance profiling plus CUDA/TensorRT kernel and optimization work. (nvidia.wd5.myworkdayjobs.com (nvidia.wd5.myworkdayjobs.com) NVIDIA has formalized production‑grade agent tooling with LangChain: the company published an AI‑Q agent blueprint on March 18, 2026 and LangChain announced an enterprise agentic AI platform built with NVIDIA, with official LangChain–NVIDIA integration packages available on GitHub. (developer.nvidia.com) (github.com) The posting also requests backgrounds in AI safety, evaluation frameworks, and agent alignment, indicating the role combines responsibility for model behavior, risk controls, and production performance in deployed agentic systems. (nvidia.wd5.myworkdayjobs.com)
