Research Engineer, Gemini Cyber Post-TrainingNew
Google DeepMind
Job AI safety & policy
- Where
- Work mode not stated: Mountain View, CA / New York, NY · USAMountain View, CA / New York, NY
- Level
- Mid career, 5–9 yrsMid5–9 yrs
- Pay
- $147k–$210k, top of range above $100k a year$147,000 – $210,000
- Apply by
- Rollingposted 27 Oct
- Hours and type
- Full-time
- Kind of work
- Technical & Engineering
- Added to CFI
- 27 October 2026
- Network
About the role
Builds reinforcement-learning environments, datasets and post-training recipes to make Gemini better at finding and fixing software vulnerabilities. Also builds evaluation pipelines and benchmarks for cyber agents.
Summary of the employer's page written by AI. Check the full listing before you apply.
Requires
PhD in CS or equivalent; 2 years in RL environments; distributed systems; vulnerability research
This role's deadline has passed, so it is no longer on the board. See open roles.