Heygen
Software Engineer, GPU Performance
Los Angeles, Palo Alto, San Francisco, Toronto, Singapore +4 · Full-time · Posted Sep 28, 2026
- Location
- Los Angeles, Palo Alto, San Francisco, Toronto, Singapore
Los Angeles, California, United States
Palo Alto, California, United States
San Francisco, California, United States
Toronto, Ontario, Canada
About the job
Optimize GPU performance across HeyGen’s AI model execution and inference infrastructure, improving latency, throughput, GPU utilization, and cost for video workloads.
What you'll do
- Profile GPU workloads, identify bottlenecks, and measure optimization impacts.
- Improve performance through batching, scheduling, memory management, and high-performance GPU kernels.
- Build performance benchmarks and regression checks, and collaborate to deploy optimizations to production.
What they ask for
- Experience optimizing GPU-based AI workloads or high-performance computing systems.
- Proficiency in Python and experience with PyTorch or a similar machine learning framework.
- Experience using profiling tools to connect hardware behavior to application bottlenecks and validate improvements.
- Ability to turn performance experiments into reliable production changes and communicate tradeoffs clearly.
Nice to have
- Experience with CUDA, Triton, or C++ GPU programming.
- Experience optimizing video, image, audio, diffusion, or Transformer models; multi-GPU inference, GPU interconnects, quantization, large-scale model serving, or performance benchmark infrastructure.
Tools
NVIDIA Nsight Systems, Nsight Compute, PyTorch Profiler, Python, PyTorch, CUDA, Triton, C++
AI focus
Optimize GPU execution and inference for AI applications, including video workloads; improve performance while measuring effects on output quality.
Model-written summary · the original posting is authoritative