OpenAI
Software Engineer, AI accelerator Runtime
$266K–$445K
San Francisco · Hybrid · Full-time · Posted Aug 24, 2026
- Visa sponsorship
- Candidates may need to meet certain legal status requirements under U.S. export control laws and regulations.
About the job
Build the low-level runtime that schedules and manages execution on OpenAI’s custom AI accelerator, working across software and hardware teams to validate and optimize it.
What you'll do
- Develop kernel scheduling, command submission, memory management, synchronization, and interfaces for the accelerator runtime.
- Use cycle-accurate simulation to develop, validate, debug, and tune runtime behavior before and after silicon availability.
- Create testing and observability tools, diagnose cross-layer issues, and collaborate on hardware-software interface improvements.
What they ask for
- Strong low-level systems programming experience in C, C++, Rust, or comparable environments.
- Experience building runtimes, drivers, firmware, operating-system components, accelerator software, or adjacent systems infrastructure.
- Understanding of concurrency, synchronization, asynchronous execution, memory-ordering, memory management, and hardware-software interfaces.
- Hands-on experience with event-based, cycle-accurate simulators or closely related architectural and performance models.
Nice to have
- Quantitative reasoning about scheduling, latency, throughput, utilization, contention, and resource tradeoffs.
- Skill debugging failures across software abstractions, device interfaces, and hardware behavior.
Pay
$266K – $445K USD base salary
Tools
C, C++, Rust
AI focus
Develop the runtime software for OpenAI’s custom AI accelerator, supporting execution of workloads for AI models.
Model-written summary · the original posting is authoritative