AI Safety & Oversight Systems Researcher

OpenAI; San Francisco (CA); Hybrid; USD 195,000 - 230,000; Full time

Benefits offered by this job

Relocation assistance

Hybrid work model

Job summary

OpenAI is seeking a safety and security minded researcher or engineer to design, build, and test practical mitigations for increasingly capable AI agents. The role focuses on oversight and system-level controls to keep agents safe in real environments, with emphasis on practical deployment and ongoing evaluation.

This role is based in San Francisco, CA, with a hybrid model of 3 days in the office per week and relocation assistance for new hires.

Qualifications

• Background in AI control or security welcome but not required.
• Ability to reason concretely about isolation boundaries, permissions, attack surfaces, and failure modes.
• Experience building robust experimental infrastructure and evaluations.

Responsibilities

• Design, build, and evaluate system-level controls for agent actions and sandboxing.
• Collaborate with Codex harness engineering to productionize AI controls.
• Red-team end-to-end agentic systems to measure data exfiltration, unsafe tool use, and harms.
• Improve the safety-productivity tradeoff by reducing missed harms and latency.

Skills

Security instincts

Threat modeling

Experimental design

Job description

OpenAI is seeking a safety and security minded researcher or engineer to design, build, and test practical mitigations for increasingly capable AI agents. The role focuses on oversight and system-level controls to keep agents safe in real environments, with emphasis on practical deployment and ongoing evaluation.

This role is based in San Francisco, CA, with a hybrid model of 3 days in the office per week and relocation assistance for new hires.