Agent Post-Training, Computer Use Research
$380k - $500k • Remote • San Francisco • FullTime
Posted 9d ago
Job Location
San Francisco
Tech Stack
Remote Work Policy
Fully remote
Employment Type
FullTime
Categories
AI Agent Engineer
About the job
The Agent Post-Training team at OpenAI is responsible for developing the frontier agents that are shipped to users. This involves training models for products like Codex and ChatGPT, focusing on creating persistent, proactive intelligence that can operate computers, collaborate with humans and other agents, and expand user capabilities. The team defines the next generation of agent abilities, builds the training signals, and runs experiments to realize these advancements. The role specifically focuses on teaching models to operate computers, including navigating browsers and desktops, using tools, reasoning through complex workflows, and completing long-horizon tasks reliably. This position is at the intersection of frontier model training, product behavior, evaluation, and systems engineering, directly influencing the computer-use capabilities of OpenAI's future agents.
Responsibilities
- Design and execute experiments to enhance agentic model behavior for complex computer use, including desktop and browser interactions.
- Manage end-to-end improvements of the post-training stack, encompassing RL, data pipelines, graders, reward signals, evals, diagnostics, and model-behavior analysis.
- Develop evaluation environments to identify model failures and convert these into training data, product fixes, or research directions.
- Collaborate with product teams to understand user needs and translate product signals into model improvements.
- Work on early-training and alignment interventions, including data mixtures, objectives, synthetic data, and eval loops.
- Contribute to decisions on which integrations, capabilities, and fixes are ready for major model runs.
- Enhance the infrastructure for large-scale training and launch, focusing on experiment velocity, reliability, observability, reproducibility, cost, latency, and production readiness.
- Undertake cross-functional projects involving model training, product infrastructure, and agent harness, such as multi-agent systems or training in production-like environments.
- Debug complex failures in models and translate qualitative behavior into concrete hypotheses, experiments, and fixes.
Requirements
- Strong technical fundamentals in machine learning, software engineering, systems, statistics, or a related field, with the ability to learn quickly.
- 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.
- Excitement for open-ended problems requiring research taste and engineering execution.
- Focus on product impact and model behavior, with opinions on agent usefulness, reliability, honesty, tastefulness, and ease of use.
- Ability to translate vague behavioral problems into concrete experiments, including hypothesis definition, pipeline building, model execution, and result analysis.
- Comfort working across research, product, infrastructure, data, evals, and safety boundaries, with clear communication skills.
- Experience building load-bearing systems and processes.
- Desire to train and ship models that make agents genuinely useful for developers, enterprises, researchers, and everyday users.