Security Engineer, Agentic Detection and Response
Google DeepMindNew York, NY
AI safety & policy
Information security
Full-time
AI safety & policy
Information security
Full-time
Salary
$147000 - $210000
Routes to impact
Skill-building & building career capital
Direct high impact on an important cause
Description
Investigate and respond to agentic AI security incidents, building detections and forensic tooling for frontier-model environments.
- Reconstruct Chain-of-Thought and tool-call trajectories to assess anomalies and possible sandbox escapes
- Coordinate live containment and blast-radius scoping across security and DeepMind response teams
- Build AI agent skills, detection rubrics, and automated forensic pipelines in Python and SQL
- Hunt for novel agentic behaviors and investigate high-ambiguity security events
- Develop detection and security tooling for rapidly evolving frontier models, with rotational on-call support
This text was generated by AI. If you notice any inconsistencies, please let us know using this form.



