Software Engineer, Accelerators

Responsibilities

• Prototype and enable OpenAI's AI software stack on new, exploratory accelerator platforms.
• Optimize large-scale model performance (LLMs, recommender systems, distributed AI workloads) for diverse hardware environments.
• Develop kernels, sharding mechanisms, and system scaling strategies tailored to emerging accelerators.
• Collaborate on optimizations at the model code level (e.g. PyTorch) and below to enhance performance on non-traditional hardware.
• Perform system-level performance modeling, debug bottlenecks, and drive end-to-end optimization.
• Work with hardware teams and vendors to evaluate alternatives to existing platforms and adapt the software stack to their architectures.
• Contribute to runtime improvements, compute/communication overlapping, and scaling efforts for frontier AI workloads.

Requirements

• 3+ years of experience working on AI infrastructure, including kernels, systems, or hardware-software co-design.
• Hands-on experience with accelerator platforms for AI at data center scale (e.g., TPUs, custom silicon, exploratory architectures).
• Strong understanding of kernels, sharding, runtime systems, or distributed scaling techniques.
• Familiarity with optimizing LLMs, CNNs, or recommender models for hardware efficiency.
• Experience with performance modeling, system debugging, and software stack adaptation for novel architectures.
• Exposure to mobile accelerators is welcome, but experience enabling data center-scale AI hardware is preferred.
• Ability to operate across multiple levels of the stack, rapidly prototype solutions, and navigate ambiguity in early hardware bring-up phases.
• Interest in shaping the future of AI compute through exploration of alternatives to mainstream accelerators.