Researcher, Agent Safety, Oversight and System Mitigations
Core
Design, build, and evaluate system-level controls and oversight mechanisms to ensure AI agents act safely and autonomously in real environments.
Role type
Senior IC researcher (AI safety, agent control, and system security)
Builds
Production-ready oversight systems, sandboxing solutions, and threat models for agentic workflows
Domain
Frontier AI safety, agent control, and system security
Deliverable
production ML models | product features
Required skills
systems security instincts, threat modeling, experimental infrastructure design, evaluation design, agent behavior analysis, isolation boundary reasoning
Preferred skills
AI control background, security background, experience with red-teaming agentic systems
Technologies
sandboxing, process isolation, permission boundaries, AI harnesses
Responsibilities
Design and build system-level controls for agent actions; Red-team end-to-end agentic systems to measure control effectiveness; Improve safety–productivity tradeoffs by reducing missed harmful actions and latency; Collaborate with engineering teams to productionize AI controls
Seniority
Senior, hands-on IC