Agent Post-Training, Computer Use Research

The role

You will teach models to operate computers by navigating browsers and desktops. You will design experiments, build evals, and improve the post-training stack to shape agentic model behavior.

What you will do

• Design and run experiments that improve agentic model behavior for complex computer use.
• Own end-to-end improvements to the post-training stack.
• Build evals and environments that expose model failures.
• Partner with Codex and ChatGPT product teams to translate product signal into model improvements.
• Work on early-training and alignment interventions.
• Help decide which integrations, capabilities, and fixes are ready for inclusion in major model runs.
• Improve the machinery for large-scale training and launch.
• Take on cross-functional projects touching model training and product infrastructure.
• Debug hard failures in shipped or near-shipped models.

Required qualifications

• Strong technical fundamentals in machine learning, software engineering, systems, statistics, or a related field.
• 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.

About the company

OpenAI is an AI research and deployment company dedicated to ensuring that general-purpose artificial intelligence benefits all of humanity. They push the boundaries of AI capabilities and seek to safely deploy them through products.