OpenAI is seeking a highly technical engineer to build the low-level device runtime for its custom AI accelerator and to optimize kernel launches, memory management, and synchronization across hardware and software boundaries.
You will work with compiler, kernel, architecture, verification, and silicon teams, using cycle-accurate simulators to validate behavior before silicon becomes available and after deployment to ensure performance and correctness.
#J-18808-LjbffrAI Accelerator Runtime Engineer (Low-Level Systems) in san francisco at Unknown Company
This position is listed as full time and onsite.