NVIDIA pioneered the GPU in 1999 and now leads the world in AI infrastructure, with its GPUs powering modern graphics, accelerated computing, and the most advanced AI systems that drive breakthroughs in generative AI, scientific discovery, autonomous machines, and massive AI‑powered data centers.
We are looking for a highly motivated and creative Staff System Software Engineer who is experienced and passionate about diagnostics for NVIDIA next-generation GPU products. You will lead and contribute to the design, implementation, and integration of these software-based validations into the sophisticated manufacturing flow for NVIDIA GPU products.
What you’ll be doing:
- Implement and enhance GPU diagnostics covering power, thermal, memory, PCIe, NVLink, and system‑level checks on boards, servers, and racks.
- Develop stress and validation tests that exercise GPU subsystems and platform components; add clear pass/fail criteria, telemetry, and error codes suitable for automation and failure attribution.
- Contribute to the integration of diagnostics into L6/L10/L11 factory flows and datacenter workflows, working with senior engineers to define coverage, runtime, and sequencing.
- Execute and monitor automated regression tests and pipelines on top of orchestration systems.
- Debug issues in cooperation with hardware, firmware, and other teams (Ops/TE/AE, etc.); root‑cause problems that span HW/FW/SW boundaries.
- Improve logging and reporting for diagnostics (e.g., structured logs, JSON, exit codes) to make triage simpler.
- Analyze factory and field data (yields, error trends, intermittency) to identify gaps in diagnostics or weak debug signals and propose targeted test or logging improvements.
- Help write and maintain technical documentation: test specs, user guides, SOPs, and troubleshooting guides for internal teams and external partners (ODMs/OEMs/CSPs).
What we need to see:
BS or MS in Computer Science, Electrical/Computer Engineering, or a related field, with 8+ years of relevant experience in system software, diagnostics, or platform validation.
Strong programming skills in C/C++ and Python, plus familiarity with shell scripting for lab tools and automation.
Experience developing low‑level system software on Linux (e.g., board bring‑up tools, diagnostics, or drivers), working close to firmware and hardware registers.
Solid understanding of server platforms:
- x86 and/or ARM server architecture (CPU, memory, PCIe topology, firmware/boot flow).
- Basic GPU architecture concepts and how GPU accelerators are integrated into servers.
Ability to read and interpret HW and system logs and understand system/block diagrams and schematics in order to connect failures back to the underlying architecture.
Demonstrated debugging and problem‑solving skills, including use of tools like gdb, perf, tracing frameworks, and vendor debug utilities.
Demonstrated technical leadership with a strong commitment to driving projects, investigations, and critical bugs to closure across cross-functional teams.
Strong communication skills and a collaborative mindset; comfortable working with globally distributed teams across HW, FW, SW, operations, and manufacturing.
Ways to stand out from the crowd:
- Hands‑on experience with diagnostic tools in factory or datacenter environments.
- Direct involvement in defining or improving diagnostic flows or RMA qualification flows for complex hardware products.
- Proven track record writing clear, high‑coverage test plans and test cases for complex HW/SW systems.
- Experience using AI‑assisted tools (for example, for log triage, code assistance, data analysis, or test generation) to accelerate diagnostics development and debugging.
NVIDIA is widely considered to be one of the technology world’s most desirable employers. We have some of the most forward‑thinking and hardworking people on the planet working for us. If you’re passionate about building and scaling world‑class diagnostics for cutting‑edge GPU servers, we want to hear from you!