Agent Post-Training, Computer Use Research

Responsibilities

• Design and run experiments to improve agentic model behavior for complex computer use, including desktop and browser navigation.
• Own end-to-end improvements to the post-training stack, including RL, data pipelines, graders, reward signals, and evals.
• Build environments and evaluations that expose model failures and translate them into training data or new research directions.
• Partner with product teams to translate user needs into model improvements for Codex and ChatGPT.
• Work on early-training and alignment interventions, including data mixtures, synthetic data, and objective functions.
• Improve large-scale training machinery, focusing on experiment velocity, reliability, and production readiness.

Requirements

• Strong technical fundamentals in machine learning, software engineering, systems, or statistics.
• Hands-on experience with Large Language Models (LLMs).
• Experience with Reinforcement Learning (RL) or RLHF/RLAIF.
• Experience with post-training techniques, evaluations, or graders.
• Experience working with synthetic data or production ML systems.

Preferred Qualifications

• Experience building coding agents or tool-using agents.
• Ability to move from vague behavioral problems to concrete, actionable experiments.
• Strong communication skills to work across research, product, and safety boundaries.

Benefits

• Competitive base salary and generous equity.
• Comprehensive medical, dental, and vision insurance.
• 401(k) retirement plan with employer match.
• Paid parental leave and flexible PTO.
• Daily meals in the office and meal delivery credits.
• Annual learning and development stipend.

About the Company

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