Research Scientist or Engineer (Agentic Systems)New
White Circle
Job AI safety & policy
- Where
- Hybrid: Hybrid · Paris / London · UKHybrid, Paris / London
- Level
- Early–Mid career, 1–9 yrsEarly–Mid1–9 yrs
- Pay
- Not listed
- Apply by
- Rollingposted 29 Jun
- Hours and type
- Full-time
- Kind of work
- Research
- Added to CFI
- 22 August 2026Posted by employer 29 June 2026
- Network
About the role
Build adversarial single- and multi-agent environments that push LLM agents to fail, and catalogue the failure modes. Findings feed internal models, product guardrails and public writeups.
Summary of the employer's page written by AI. Check the full listing before you apply.
Requires
Built at least one non-trivial agent environment or research pipeline; empirical research record
This role's deadline has passed, so it is no longer on the board. See open roles.