Frontier Agent, Post-Training: Computer-Use Research

OpenAI; San Francisco (CA); On-site; USD 380,000 - 500,000; Full time

Job summary

OpenAI's Agent Post-Training, Computer Use team is seeking a contributor to teach models to operate computers, navigate browsers and desktops, use tools, and complete long-horizon tasks with reliability and judgment.

You will collaborate with researchers, engineers, product and infrastructure teams to decide what goes into major model runs, measure outcomes, and ship improvements into products used by real people. This high-autonomy role directly impacts the next generation of frontier agents.

Qualifications

• Strong technical fundamentals in machine learning, software engineering, systems, and statistics.
• Hands-on experience with large language models, RL, RLHF/RLAIF, post-training, evals, graders, or synthetic data is a plus.
• Ability to translate research ideas into engineering experiments and product improvements.

Responsibilities

• Design and run experiments that improve agentic model behavior for complex computer use, including desktop and browser.
• Own end-to-end improvements to the post-training stack, including RL, data pipelines, graders, reward signals, evals, diagnostics, and model-behavior analysis.
• Build evals and environments that expose the next set of model failures and turn them into training data or product fixes.

Skills

Machine learning fundamentals

Software engineering

Systems

Statistics

LLMs/RL/RLHF experience

Job description

OpenAI's Agent Post-Training, Computer Use team is seeking a contributor to teach models to operate computers, navigate browsers and desktops, use tools, and complete long-horizon tasks with reliability and judgment.

You will collaborate with researchers, engineers, product and infrastructure teams to decide what goes into major model runs, measure outcomes, and ship improvements into products used by real people. This high-autonomy role directly impacts the next generation of frontier agents.