jobs

Researcher, Agent Safety, Training and Evaluations

About the TeamThe Agent Safety team works to ensure that increasingly capable AI agents act safely, exercise sound judgment, and remain aligned with user intent. Our mission is to reduce the probability of severe unintended outcomes from in

In inglese

Company

OpenAI

Location

San Francisco

Status

Open

Posted

2026-09-03T16:33:13.278+00:00

OpenAI's Agent Safety team is hiring a researcher to train and evaluate frontier models so agents act safely and follow user intent. Work spans training methods, environments and data; evaluations and production metrics that flag emerging risks; and oversight mechanisms. The person mines incidents, builds scalable measurement and evaluation systems, and works with post-training, capabilities, oversight and pre-training teams. The posting asks for strength in research or ML engineering, quantitative or applied model research, and comfort owning ambiguous projects. Prior safety experience is not required.