Head of Safety

Highlights

• Own end-to-end safety strategy for Devin, a capable autonomous coding agent.
• Design evals and red-teaming for prompt injection, exfiltration, and escape.
• Set deployment policy and oversight for agent actions with enterprise customers.

Ideal candidate

Hands-on safety leader from a frontier lab who has shipped real mitigations and wants to secure autonomous agents in production. Builder mindset with strong judgment and drive to own outcomes end to end.

Requirements

• Hands-on safety, alignment, evaluations, or red-teaming work at a frontier AI lab
• Shipped safety work that affected real models or products
• Deep understanding of agentic failure modes and mitigation approaches
• Track record of published or widely used safety research, evals, or methods
• Technical fluency in Python and training and inference stacks
• Experience starting or scaling a safety function focused on shipping mitigations
• Strong judgment on deployment risk under uncertainty