Description
OpenAI is hiring a Hardware Optimization and Co-Design Engineer to co-design future hardware from vendors for programmability and performance, support optimal kernel development, develop performance estimates and system models, and influence hardware and networking roadmaps for efficient large-language-model training and inference. The role requires 4+ years of industry experience, software/hardware co-design expertise, GPU or AI accelerator knowledge, CUDA or Triton programming experience, low-precision ML optimization experience, and strong C/C++ and Python skills. It is based in San Francisco, California, with a hybrid model of three office days per week and relocation assistance.
