Member of Technical Staff GPU Performance Engineer

You will write and integrate custom GPU kernels, profile training and inference workflows, and remove performance bottlenecks. You will build correctness and numerical tests, maintain benchmark guardrails, and turn research ideas into reliable production speedups.

Responsibilities

• Write high-performance GPU kernels
• Integrate kernels into PyTorch pipelines
• Profile and optimize training and inference workflows
• Build correctness tests and numerical checks
• Build and maintain performance benchmarks and regression guardrails
• Turn research ideas into shipped performance improvements

Requirements

• Experience authoring custom CUDA kernels
• Understanding of GPU architecture and performance, including memory hierarchy, warps, shared memory, register pressure, bandwidth, and compute limits
• Proficiency with Nsight Systems or Nsight Compute and performance methodology
• Strong C and C++ skills

Benefits

• Equity
• Medical, dental, and vision premiums fully paid for employees and dependents
• 401(k) matching up to 4% of base pay
• Unlimited PTO
• Company-wide Refill Days