Genmo
GPU Performance Engineer
San Francisco HQ · Full-time · Posted Jul 17, 2025
- Experience
- 5+ years systems programming experience, with 3+ years focused on GPU optimization.
- Education
- Bachelor's or Master's degree in Computer Science, Electrical Engineering, or related field.
About the job
Optimizes GPU workloads and model-serving infrastructure, including kernel performance and cold-start latency.
What you'll do
- Profile GPU workloads and implement performance improvements across the stack.
- Develop CUDA and Triton kernels; tune memory access, kernel fusion, and GPU utilization.
- Reduce serving cold-start latency and collaborate with ML engineers on model optimization.
What they ask for
- Bachelor's or Master's degree in Computer Science, Electrical Engineering, or a related field.
- 5+ years of systems programming experience, including 3+ years focused on GPU optimization.
- Expert proficiency with Nsight Systems and nvprof; strong CUDA skills with production kernel development.
- Deep understanding of GPU architecture; production experience with Python and C++.
Nice to have
- Experience with Triton, CUTLASS or similar libraries, or ML-specific optimizations.
- RDMA/InfiniBand experience, GPU library/framework contributions, or PTX/SASS debugging skills.
Tools
H100, Nsight Systems, nvprof, CUDA, Triton, CUTLASS, Python, C++
AI focus
Optimizes GPU performance for model serving and ML workloads, including attention and transformer optimizations.
Model-written summary · the original posting is authoritative