OpenAI
Software Engineer, TPU Compilers
$380K–$500K
San Francisco · Hybrid · Full-time · Posted Oct 20, 2026
About the job
Optimize OpenAI’s frontier models on TPUs, owning performance across model code, kernels, and hardware execution.
What you'll do
- Profile TPU workloads and improve latency, throughput, and hardware utilization.
- Develop and optimize kernels, and explore quantization to improve inference efficiency while preserving model quality.
- Build benchmarks and validation tools, and partner with researchers and engineers to support new architectures and production improvements.
What they ask for
- Practical experience optimizing machine learning workloads on TPUs through training, inference, or both.
- Understanding of modern model architectures and how their computation maps onto accelerator hardware.
- Strong software engineering skills and a systematic approach to profiling, debugging, and validating performance.
- Ability to connect kernel behavior to whole-model performance and work across the stack.
Nice to have
- Experience with quantization.
- Experience with attention kernels.
Pay
$380K – $500K USD base salary
Tools
TPUs
AI focus
Optimize frontier-model execution on TPUs, including inference efficiency and performance across the model forward pass.
Model-written summary · the original posting is authoritative