OpenAI
Software Engineer, Accelerators
$295K–$380K
San Francisco · Full-time · Posted Jun 27, 2025
- Experience
- 3+ years.
- Visa sponsorship
- Candidates may need to meet certain legal status requirements under U.S. export control laws and regulations.
About the job
Evaluates and enables emerging accelerator platforms for large-scale AI workloads, adapting OpenAI’s software stack and improving system performance.
What you'll do
- Prototype OpenAI’s AI software stack on exploratory accelerator platforms and optimize large-scale model performance across hardware environments.
- Develop kernels, sharding mechanisms, and scaling strategies; model performance, debug bottlenecks, and drive end-to-end optimization.
- Collaborate with hardware teams and vendors to evaluate platforms and improve runtimes, compute/communication overlap, and workload scaling.
What they ask for
- 3+ years of experience working on AI infrastructure, including kernels, systems, or hardware-software co-design.
- Hands-on experience with AI accelerator platforms at data center scale.
- Strong understanding of kernels, sharding, runtime systems, or distributed scaling techniques.
- Experience with performance modeling, system debugging, and software-stack adaptation for novel architectures.
Nice to have
- Familiarity with optimizing LLMs, CNNs, or recommender models for hardware efficiency.
- Exposure to mobile accelerators; data center-scale AI hardware enablement is preferred.
Pay
$295K – $380K USD base salary
Tools
PyTorch, TPUs
AI focus
Enables and optimizes AI training and inference workloads on emerging accelerators, including LLMs, CNNs, recommender systems, and distributed AI workloads.
Model-written summary · the original posting is authoritative