- Where
- On-site: On-site · San Francisco, CA · USAOn-site, San Francisco, CA
- Level
- Entry–Early career, 0–4 yrsEntry–Early0–4 yrs
- Pay
- Not listed
- Apply by
- Rollingposted 14 Jul 2025
- Hours and type
- Full-time
- Kind of work
- Technical & Engineering
- Added to CFI
- 22 August 2026Posted by employer 14 July 2025
- Network
About the role
The engineer designs and optimises high-performance GPU kernels for training and inference at an AI interpretability lab. They work with ML researchers to build performance-aware models.
Summary of the employer's page written by AI. Check the full listing before you apply.
Requires
Demonstrated ML kernel work; PyTorch and Triton/TileLang familiarity
This role's deadline has passed, so it is no longer on the board. See open roles.