Researcher, Agent Safety, Training and Evaluations
San Francisco
Workplace: HybridFull timeUSD 380,000 - 500,000 annuallyFunction: Research & Scientific (R&D)Skills: ["Judgment","Comfort with ambiguity","Independent execution","Collaboration","Technical execution"]Work on OpenAI’s Agent Safety mission by training and evaluating frontier AI agents to reduce harmful or misaligned actions. Build measurement, data-processing, and evaluation systems that convert real incidents into repeatable safety signals. Collaborate with training, capabilities, oversight, and pre-training partners to ship research-backed mitigations into large-scale agent systems.
Loading
Loading job details...
Preparing the role view and application actions.

