Frontier AI Evaluation & Environments Researcher

Agent Post-Training, Frontier Evals and Environments Research; San Francisco (CA); On-site; USD 180,000 - 240,000; Full time

Job summary

OpenAI is seeking a researcher for Frontier Evals & Environments to help build north star model environments driving safe AGI/ASI progress. You’ll shape the research programs behind OpenAI’s most ambitious training runs and influence data, environments, graders, and feedback loops.

You will collaborate with researchers, engineers, product and safety teams to decide what to include in major model runs, measure outcomes, and ship improvements to products used by people.

Qualifications

• Strong technical fundamentals in machine learning, software engineering, systems, or statistics.
• Hands-on experience with LLMs, RL, RLHF/RLAIF, post-training, evals, graders, synthetic data, model training, coding agents, tool-using agents, or production ML systems.

Responsibilities

• Create ambitious RL environments to push models to their limits and measure frontier model capabilities, skills, and behaviors.
• Develop new methodologies for automatically exploring the behavior of these models.
• Dive deep into the science of measurement, including scalability, reliability, and variance of evaluation methodology.
• Help steer training for our largest training runs and see the future first.
• Design scalable systems and processes to support continuous evaluation.
• Build self-improvement loops to automate model understanding.

Skills

ML fundamentals

Software engineering

Systems

Statistics

LLMs

RL basics

Post-training

Evaluations

Graders

Synthetic data

Model training

Agents tooling

Production ML

Job description

OpenAI is seeking a researcher for Frontier Evals & Environments to help build north star model environments driving safe AGI/ASI progress. You’ll shape the research programs behind OpenAI’s most ambitious training runs and influence data, environments, graders, and feedback loops.

You will collaborate with researchers, engineers, product and safety teams to decide what to include in major model runs, measure outcomes, and ship improvements to products used by people.