openai
Software Engineer, Hardware
San Franciscofulltimemid
About this role
OpenAI seeks a Software Engineer for its Hardware team to build and optimize the low-level software stack that orchestrates computation across AI supercomputing clusters. You'll design runtimes, develop custom kernels, contribute to compiler infrastructure, and create simulation systems to validate distributed training workloads.
What you'll do
- Design APIs and runtime components for orchestrating computation and data movement across ML workloads
- Contribute to compiler infrastructure including optimizations and compiler passes
- Engineer and optimize compute kernels for correctness and high performance
- Profile and optimize system bottlenecks in I/O, memory, and interconnects
- Develop simulation infrastructure to validate runtime behaviors and support hardware development
- Deploy runtime and compiler updates to supercomputing builds in collaboration with hardware teams
What they're looking for
- Systems programming (Rust, C++)
- Python scripting
- Compiler development
- High-performance computing
- Distributed systems
- Runtime systems or accelerator programming
- Kernel authoring
- ML infrastructure knowledge
Benefits
- Hybrid work model (3 days/week in-office)
- San Francisco location
- Relocation assistance available
- Equal opportunity employer
- Opportunity to influence training framework architecture
Opens the official application on the employer’s site. No login required.