Research Engineer, Frontier Evals & Environments

Highlights

• Build north-star RL environments guiding OpenAI's largest training runs.
• Measure frontier agent capabilities and steer models into shipped products.
• Design continuous evaluation systems and self-improvement loops for models.

What they offer

• Equity

Ideal candidate

High-agency researcher-engineer excited by open-ended measurement problems who cares about useful, reliable agents and wants hands-on influence over frontier training runs and shipped products.

Requirements

• Strong technical fundamentals in ML, software engineering, systems, statistics or related field
• Hands-on experience with LLMs, RL, post-training, evals, graders, synthetic data or agents
• Excited by open-ended problems with noisy signal requiring research taste and execution
• Care about product impact and useful, reliable, honest agent behavior
• Can move from vague behavioral problem to concrete experiment and decision
• Comfortable working across research, product, infrastructure, data and safety
• Like building load-bearing systems and processes when the team needs it