Principal Software Engineer, GPU Firmware and GPU System Software — CSP Engagements
Core
Principal Software Engineer acting as technical focal point for GPU firmware and system software, ensuring CSP/hyperscale customers can reliably manage, update, and operate NVIDIA GPUs at fleet scale.
Role type
Principal Software Engineer (GPU Firmware & System Software)
Builds
Shared understanding of GPU firmware integration, customer-side automation/recovery procedures, and fleet-scale update orchestration.
Domain
Semiconductor / GPU Architecture / Firmware Engineering
Deliverable
production ML models | product features | infrastructure
Required skills
GPU system software engineering, GPU firmware engineering, accelerator platform engineering, GPU architecture internals (SMs, GEMM, memory hierarchy), multi-GPU fabric architectures (NVLink), firmware update lifecycle management, GPU error handling and recovery flows, GPU health monitoring and telemetry
Preferred skills
NVIDIA GPU VBIOS and microcontroller firmware experience, GPU fleet management at 10K+ scale, GPU error taxonomy expertise, GPU security (secure boot, attestation), GPU power management architecture
Technologies
VBIOS, InfoROM, microcontroller firmware, NVLink, Xid errors, ECC counters
Responsibilities
Drive GPU firmware work streams with CSP engineering teams, gather and synthesize CSP feedback into NVIDIA's roadmap, drive GPU firmware update orchestration for large-scale deployments, serve as technical focal point for GPU behaviors documentation, identify cross-CSP GPU SW/FW issue patterns
Seniority
Principal, hands-on IC with strategy & mentorship
