- Palo Alto, California, United States +1 more
- On-site
- Full-time · Staff
Founding engineer to own kernel-level performance across the inference stack, identifying bottlenecks, writing and fusing GPU kernels, and using agents to automate kernel optimization.
Locations (2)
- Palo Alto, California, United States
- Brooklyn, New York, United States
Compensation
Highly competitive salary and equity
Responsibilities
- Identify bottlenecks in kernel-level performance
- Write and fuse GPU kernels
- Use agents to automate kernel optimization
- Carry improvements through to production
- Define projects from first principles
Requirements
- Nice to have: Custom accelerators
- Nice to have: CUTLASS
- Nice to have: cuBLAS
- Nice to have: NCCL
- Nice to have: CuTe DSL
