- Where
- Work mode not stated: London / Berkeley · USA, UKLondon / Berkeley
- Level
- Mid–Senior career, 5–10+ yrsMid–Senior5–10+ yrs
- Pay
- £120k–£160k, top of range above $100k a year£120,000 – £160,000
- Apply by
- Rollingposted 1 Jul
- Hours and type
- Full-time
- Kind of work
- Research
- Added to CFI
- 22 August 2026Posted by employer 1 July 2026
- Network
- Not listed
About the role
Design and run experiments on LLM-based agents to test theoretically predicted risks such as deception and power-seeking. Build and evaluate safety mitigations and help shape the research agenda of a three-person AI safety non-profit.
Summary of the employer's page written by AI. Check the full listing before you apply.
Requires
Maths background to learn AIXI and RL theory; experience running LLM-agent experiments
This role's deadline has passed, so it is no longer on the board. See open roles.