jobs

Research Engineer, Frontier Evals & Environments

About the teamThe Agent Post-Training team creates the frontier agents OpenAI ships to the world. We are training the models behind our agents in Codex, ChatGPT, the API, and other frontier products: persistent, proactive intelligence that

In inglese

Company

OpenAI

Location

San Francisco

Status

Open

Posted

2025-04-13T17:25:42.666+00:00

Research engineer on OpenAI's Agent Post-Training team, building evaluation environments and benchmarks that steer the company's largest training runs. Work includes designing reinforcement-learning environments, methods that automatically probe model behavior, measurement research on reliability and variance, and scalable evaluation systems. Public benchmarks from this kind of role include SWE-bench Verified, MLE-bench and PaperBench. The posting asks for grounding in machine learning, software engineering, systems or statistics, plus hands-on experience with LLMs, post-training, evals or agents. San Francisco, on-site.