Senior Systems Software Engineer - Fleet Debuggability

NVIDIA•Santa Clara, CA

2d•$184,000 - $356,500

About The Position

NVIDIA's invention of the GPU in 1999 sparked the growth of the PC gaming market, redefined modern computer graphics, and revolutionized parallel computing. More recently, GPU deep learning ignited modern deep learning — the next era of computing — with the GPU acting as the brain of computers, robots, and self-driving cars that can perceive and understand the world. Today, we are increasingly known as “the AI computing company.” We're looking to grow our company and establish teams with the most thoughtful people in the world. We are the Datacenter System Software team, and we are looking for a highly motivated, creative Senior Engineer to drive Fleet Scale Debuggability end to end. You will design, architect, and build infrastructure, tooling, analytics on how to collect multi-rack scale logs. The solution should normalize, correlate, and reason over logs spanning multiple components, trays, or racks including NVIDIA's GPUs, CPUs, Network products. The logs shall be fetched inband or out of band and should help triage fleet level issues seen by our customers. Your work directly shortens the path from a raw, noisy log stream to an actionable root cause. Join us at the forefront of technological advancement.

Requirements

10+ years in the software industry with specialization in system software and/or firmware development.
BS, MS, or PhD in CS, CE, EE, or a related technical field — or equivalent experience.
Proven track record of shipping scalable server products or fleet-wide experience.
A self-starter who loves finding creative solutions to complicated problems, with excellent written and oral communication skills — including executive-level reporting — strong work ethic, and dedication to teamwork.
Flexibility to work and communicate effectively across teams, partners, and time zones.
Experience with SCM (e.g., Git, Perforce) and project-management tools like Jira.
Strong, demonstrable skills in Python or RUST.
Deep Linux systems experience: kernel and driver logs, syslog, journald, and the realities of debugging on server platforms.
Hands-on experience with out-of-band management and platform interfaces — BMC, Redfish, IPMI, SEL — and an understanding of in-band vs. out-of-band trade-offs.
Strong skills in log parsing, normalization, and structured logging, and comfort designing schemas and taxonomies for machine-readable events.

Nice To Haves

Experience leading debuggability solutions on sophisticated rack-scale compute architectures like GB200/GB300 NVL72.
Familiarity with log and telemetry analytics stacks (e.g., OpenSearch/ELK, Loki, Prometheus, Grafana, PagerDuty) and time-series databases.
Hands-on experience with x86/ARM system architecture and coding (C/C++, Python).
Experience with SCM (Git, Perforce) and project management tools (Jira).
Track record of integrating AI/LLM tooling into engineering workflows — for triage, validation, log analysis, or test generation.
Experience standing up follow-the-sun support organizations with measurable response SLAs
Experience contributing to or maintaining open-source projects, including managing the boundary between internal and public code.

Responsibilities

Architect, Design, build, fleet-wide log collection and analysis solutions that aggregate signals across components, trays, and racks.
Develop tooling to collect, normalize, and time-align logs from heterogeneous sources — kernel and driver logs, syslog, Redfish event logs, SEL, firmware and BMC logs — over both in-band and out-of-band channels.
Build and maintain a log catalog and taxonomy that maps raw log signatures to fault classes, severity, and remediation guidance, so triage is repeatable rather than tribal knowledge.
Develop debug and root-cause tooling that turns high-volume fleet logs into ranked, actionable diagnoses for hardware, firmware, and platform faults.
Drive the design for collecting and analyzing logs at fleet scale while keeping overhead on production compute nodes low.
Partner with all matrixed organizations — developers, SWQA, and product engineering — in a fast-moving environment with end-to-end logging solutions, event schemas, and the contract between log producers and your tooling.
Steward the project's open-source release: keep internal and public code paths clean, review community contributions, and represent the tooling in upstream discussions.
Write design docs and own end-to-end delivery, working across teams from definition through implementation, debugging, testing, and early customer support.
Perform code reviews and partner with development and QA to strengthen unit testing, integration coverage, and test plans.
Track work through Jira and bug-management tools and build a realistic end-to-end execution plan in collaboration with other engineers and managers.