ML Silicon Performance Lead
OpenAI · San Francisco
Frontier labonsitelead$287k–$485k
At a glance
Highlights
- Drive HW/SW co-design shaping how ML workloads map to AI-native silicon.
- Guide next-gen AI silicon for frontier models via modeling and simulation.
- Define specialized acceleration IP for large-scale AI workloads.
What they offer
- equity
Ideal candidate
Systems-minded accelerator expert who loves connecting ML workloads to silicon design, thriving in ambiguity and driving alignment across research, compiler, and hardware teams.
Requirements
- Deep experience with ML accelerator architecture and how compute, memory, and data movement shape performance
- Ability to translate model-level behavior into kernel requirements and hardware design decisions
- Experience building or using accelerator performance models and simulators
- Hands-on experience optimizing ML kernels for utilization, locality, and execution efficiency
- Understanding of quantization and numerical formats and implications for model quality and hardware
- Experience with hardware/software co-design across kernels, compilers, ISA, and dedicated hardware
- Ability to work across research, compiler, and silicon teams to align on ambiguous requirements
Lab profile
About OpenAI
Builds safe frontier AI models and products like ChatGPT for broad benefit.
What they build
- ChatGPT
- OpenAI API
- Frontier AI models
Known for
- Frontier-scale distributed ML training
- Massive GPU fleet and data center operations
- Exabyte-scale data lakes and streaming
- AI alignment and safety research
Key facts
- Operates one of the world's largest GPU fleets
Posted · Last seen