Senior GPU Systems Performance Architect

NVIDIA Corporation; Santa Clara (CA); Hybrid; USD 184,000 - 357,000; Full time

Benefits offered by this job

Equity compensation

Benefits package

Job summary

NVIDIA Corporation in Santa Clara, CA, seeks a Senior Systems Performance Engineer to build profiling infrastructure that measures the interaction between design graphs in system memory and GPU kernels.

You will benchmark EDA applications, profile CUDA and GPU bottlenecks, and develop tools for memory usage in multi-terabyte builds. This role requires 8+ years in systems performance analysis and experience with Slurm, LSF, or Kubernetes.

Qualifications

• Experience with CUDA profiling and GPU profiling tools to address PCIe bottlenecks and kernel stalls.
• Strong knowledge of profiling tools (perf, eBPF, VTune, Valgrind) and their internals.
• Experience with distributed compute environments (Slurm, LSF, Kubernetes).

Responsibilities

• Architecting and maintaining custom profiling frameworks across CPU and GPU environments.
• Conducting deep-dive benchmarking of EDA workloads to characterize memory access and occupancy.
• Using GPU profilers to detect GPU-side inefficiencies such as warp divergence and bottlenecks.
• Developing tools to monitor high-watermark memory usage in massive EDA builds.
• Developing predictive models to guide hardware procurement and cloud instance selection.

Skills

CUDA

Nsight Systems/Compute

Perf

eBPF

VTune

Valgrind

Slurm

Kubernetes

GPU profiling

System performance

Education

BS/MS/PhD in CS/EE or related

Tools

NVIDIA Nsight Systems/Compute

Job description

NVIDIA Corporation in Santa Clara, CA, seeks a Senior Systems Performance Engineer to build profiling infrastructure that measures the interaction between design graphs in system memory and GPU kernels.

You will benchmark EDA applications, profile CUDA and GPU bottlenecks, and develop tools for memory usage in multi-terabyte builds. This role requires 8+ years in systems performance analysis and experience with Slurm, LSF, or Kubernetes.