Software Engineer, TPU Compilers

OpenAI • San Francisco, CA

1d • $380,000 - $500,000

About The Position

OpenAI is seeking an engineer to lead optimization of its frontier models on TPUs. The role owns performance across the full forward pass, spanning model code, kernels, and hardware execution. The engineer identifies performance bottlenecks, selects high-impact improvements, and implements and validates them. Relevant experience may come from either model training or inference, with a strong foundation in optimizing real TPU workloads.

The role is part of the Inference team, which brings OpenAI’s research and technology to consumers, enterprises, and developers through its products. The team focuses on performant and efficient model inference and on accelerating research progression through inference.

Requirements

• 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.
• Ability to connect detailed kernel behavior to complete-model performance and work across these layers of the stack.
• Strong software engineering skills and a systematic approach to profiling, debugging, and validating performance improvements.
• Ability to take ownership of open-ended technical problems, identify priorities, make design decisions, and deliver working solutions.
• Enjoyment of collaboration with researchers and engineers, learning unfamiliar parts of a system, and communicating technical tradeoffs.

Nice To Haves

• Experience with quantization or attention kernels.

Responsibilities

• Optimize model execution on TPUs across the complete forward pass and the interactions between model code, kernels, and hardware.
• Profile workloads, identify bottlenecks, and implement improvements to latency, throughput, and hardware utilization.
• Develop and optimize kernels, using knowledge of the surrounding model to guide implementation choices.
• Explore techniques such as quantization to improve inference efficiency while preserving model quality.
• Build benchmarks and tools to measure performance, validate correctness, and guide optimization work.
• Partner with researchers and engineers to support new model architectures and bring performance improvements into production.

What This Job Offers

• Job Type: Full-time
• Education Level: No Education Listed