Software Engineer, AI accelerator Runtime
San Francisco
Workplace: HybridFull timeUSD 266,000 - 445,000 annuallyFunction: Software EngineeringSkills: ["Debugging","Collaboration","Quantitative reasoning","Problem-solving"]Build OpenAI’s low-level device runtime for custom AI accelerator hardware, turning compiled programs into efficient execution. You’ll implement kernel-launch scheduling, command submission, dependency tracking, and robust memory management across concurrent workloads. Partner with compiler, kernel, architecture, and silicon teams to define interfaces and validate behavior using event-based, cycle-accurate simulators, then diagnose correctness and performance issues. Add tests, tracing, profiling, and observability to ensure production-ready runtime capabilities.
Loading
Loading job details...
Preparing the role view and application actions.

