SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
OpenAI's Hardware organization is seeking a Software Engineer to build the low-level device runtime for custom AI accelerators. This role sits at the intersection of software and hardware, where you'll design and implement runtime systems that transform compiled programs into efficient, functional execution on OpenAI's silicon.
Key responsibilities include:
- Design and implement low-level device runtime for custom silicon
- Build kernel-launch scheduling, command submission, queueing, dependency tracking, and completion handling
- Manage device memory spaces, allocation, virtual-to-physical mappings, data movement, and lifetime across concurrent workloads
- Implement synchronization primitives, events, barriers, streams, and ordering guarantees
- Define clean interfaces between runtime, drivers, firmware, compiler-generated code, kernels, and higher-level execution systems
- Use event-based, cycle-accurate simulators to develop, validate, debug, and performance-tune runtime behavior
- Diagnose concurrency, memory-ordering, deadlock, race, correctness, and performance issues across software-hardware boundaries
- Build tests, tracing, profiling, observability, and reproducible workloads
- Partner with architecture and silicon teams to drive hardware-software interface improvements
Ideal candidates have strong low-level systems programming experience in C, C++, Rust, or comparable languages. You should have built runtimes, drivers, firmware, OS components, or accelerator software. Deep understanding of concurrency, synchronization, memory management, DMA, caching, coherency, and hardware-software interfaces is essential. Experience with cycle-accurate simulators, quantitative reasoning about scheduling and performance tradeoffs, and ability to debug across software abstractions and hardware behavior are highly valued. You'll work effectively across compiler, kernel, architecture, verification, firmware, and silicon teams.
