Senior GPU/CPU Profiling Architect

NVIDIA AI; Durham (NC); On-site; USD 224,000 - 357,000; Full time

Benefits offered by this job

Equity

Benefits package

Job summary

NVIDIA is seeking a Senior Systems Performance Engineer to build profiling infrastructure spanning CPU and GPU workloads. You will measure, analyze, and optimize interactions between large design graphs in memory and high-throughput GPU kernels.

You will drive benchmarking across multi-node environments, identify bottlenecks, and provide actionable roadmaps for kernel optimization, memory usage, and system-level efficiency.

Qualifications

• BS/MS/PhD in Computer Science, Electrical Engineering, or a related field (or equivalent experience).
• 8+ years of relevant experience with systems-level performance analysis.
• Strong profiling and benchmarking discipline across CPU and GPU environments.
• Experience with distributed compute environments such as Slurm, LSF, or Kubernetes.

Responsibilities

• Architect and maintain profiling frameworks across CPU (multi-core/multi-socket) and GPU (multi-node/NVLink) environments.
• Conduct deep-dive benchmarking of EDA workloads to characterize memory access patterns and occupancy.
• Use GPU profilers to detect GPU-side inefficiencies like warp divergence and kernel stalls.
• Develop tools to monitor and attribute high-watermark memory usage in large builds.
• Create predictive models to guide hardware procurement and cloud instance selection.

Skills

Profiling

Benchmarking

Performance analysis

Multi-node profiling

Education

BS/MS/PhD in CS/EE or related field

8+ years of relevant experience

Tools

CUDA

NVIDIA Nsight Systems/Compute

perf

eBPF

VTune

Valgrind

Slurm

LSF

Kubernetes

Job description

NVIDIA is seeking a Senior Systems Performance Engineer to build profiling infrastructure spanning CPU and GPU workloads. You will measure, analyze, and optimize interactions between large design graphs in memory and high-throughput GPU kernels.

You will drive benchmarking across multi-node environments, identify bottlenecks, and provide actionable roadmaps for kernel optimization, memory usage, and system-level efficiency.