AI safety & policy
Full-time
Routes to impact
Direct high impact on an important cause
Description
Lead safety for an autonomous AI agent, owning strategy, evaluations, deployment policies, and the safety team.
- Design evals and red-teaming for misuse, prompt injection, data exfiltration, sandbox escape, and reward hacking
- Partner with training and agent teams to build safety into models, planning, actions, and oversight
- Define deployment policies, permissions, and oversight for enterprise use
- Hire and lead safety researchers and engineers
- Represent safety work with customers, policymakers, and the research community
This text was generated by AI. If you notice any inconsistencies, please let us know using this form.



