Agent Post-Training: Computer-Use Architect

OpenAI; San Francisco (CA); On-site; USD 130,000 - 170,000; Full time

Job summary

OpenAI is looking for a member of the Agent Post-Training team in San Francisco. In this role, you will teach models to operate computers and improve model behavior for complex tasks. You will work alongside researchers and engineers to shape the next generation of agents.

This role demands strong technical fundamentals in areas like machine learning and software engineering, along with hands-on experience with relevant technologies. OpenAI is dedicated to deploying AI that benefits humanity and ensures equity in their hiring process.

Qualifications

• Strong fundamentals in machine learning, software engineering, and systems.
• Hands-on experience with LLMs, RL, RLHF/RLAIF, and model training.
• Excited about open-ended problems and product impact.

Responsibilities

• Design experiments to improve model behavior for complex tasks.
• Own improvements to the post-training stack and analysis.
• Partner with product teams to translate user needs into model enhancements.

Skills

Machine learning

Software engineering

Systems

Statistics

LLMs

Tool-using agents

Research

Education

Degree in a related field

Tools

Python

Data pipelines

Experimental design

Job description

OpenAI is looking for a member of the Agent Post-Training team in San Francisco. In this role, you will teach models to operate computers and improve model behavior for complex tasks. You will work alongside researchers and engineers to shape the next generation of agents.

This role demands strong technical fundamentals in areas like machine learning and software engineering, along with hands-on experience with relevant technologies. OpenAI is dedicated to deploying AI that benefits humanity and ensures equity in their hiring process.