jobs

Software Engineer, GPT Infrastructure

About the TeamThe GPT Infrastructure team builds systems that turn advances in model inference and optimization into reliable production capabilities. We enable OpenAI workloads to be qualified and optimized across new accelerator platforms

Company

OpenAI

Location

San Francisco; Seattle

Status

Open

Posted

2026-04-27T19:55:27.406+00:00

OpenAI's GPT Infrastructure team is hiring a software engineer to build the platform that qualifies and optimizes inference workloads across different accelerator hardware. The work covers control-plane services for long-running optimization campaigns, secure partner-side software that compiles, runs, verifies and benchmarks candidate kernels, and evaluation systems for correctness and performance. The posting asks for distributed-systems experience, proficiency in Python, C++, Go or Rust, and Linux and networking knowledge; familiarity with CUDA, ROCm, Triton, LLVM or MLIR and serving engines such as vLLM is preferred.