$272,000 - $431,250 USD yearly
Originally posted July 28, 2026 by the employer. Relisted by the employer on November 7, 2026.
About the role
As a Principal Engineer on the CSP Engagements team, you will act as the technical focal point for end-to-end performance, working directly with engineering teams of key CSP and hyperscale customers to achieve performance targets on NVIDIA platforms.
What you'll do
- Drive performance characterization work streams with key CSP/hyperscale engineering teams regarding platform performance expectations, profiling methodology, and workload-specific tuning options.
- Gather and synthesize CSP performance feedback, identify throughput gaps, and champion optimization priorities into NVIDIA CUDA, NCCL, driver, and firmware teams.
- Update and validate open-source performance and stress tools (e.g., STREAM, GPU Burn, GPU BLAST) for NVIDIA rack-scale systems, GPU architectures, and CPU platforms.
- Work with CSPs to ensure internal and external performance and validation tooling reflects GPU capabilities, memory hierarchy changes, and platform-specific tuning parameters.
- Conduct cross-CSP performance comparison and pattern analysis to resolve configuration, software, or workload differences across deployments.
- Define test strategies and tooling requirements for performance validation across NVIDIA internal certification and customer acceptance milestones.
What you'll need
- 15+ years of experience in systems performance engineering, ideally in GPU/HPC/ML infrastructure, with a BS or MS in Computer Science, Computer Engineering, or related field (or equivalent experience).
- Proficiency in GPU workload profiling using Nsight Systems, Nsight Compute, DCGM metrics, or equivalent instrumentation.
- Understanding of distributed training performance dynamics, including computation/communication overlap, pipeline bubbles, memory bandwidth utilization, and collective efficiency.
- Knowledge of statistical methods for performance analysis, including regression detection, confidence intervals, and A/B comparison at scale.
- Understanding of how the full software stack impacts performance, including driver overhead, collective algorithm selection, memory allocation, scheduling, and firmware power management.
- Data analysis and visualization skills using Python, pandas, and dashboards.
Skills: performance engineering, GPU workload profiling, distributed training, software stack optimization, CSP/hyperscale customers
More semiconductor jobs: NVIDIA jobs
Tracked in our index since 27 Jun 2026. This is our observation date, not the employer posting date. Specialty: Infrastructure/Platform Software (GPU Software Stack). 341 active roles in this specialty at 26 employers in our index, as of 21 Nov 2026.
How we counted: postings read from employer career sites. How we count