Researcher, Agentic Post-training

OpenAI

San Francisco, California, United States
Not specified; not specified; not specified
Strong ml fundamentals and llm experience
Hands-on experience with rl and rlhf
Ability to own ambiguous problems end-to-end
This role involves owning end-to-end research and engineering projects to improve the post-training of OpenAI's agentic models used by hundreds of millions of people

Job Summary

  • This role involves owning end-to-end research and engineering projects to improve the post-training of OpenAI's agentic models used by hundreds of millions of people.
  • The team is responsible for developing horizontal improvements such as factuality, instruction following, tool use, and multi-agent collaboration.
  • Candidates must be deeply technical engineers who can move quickly in complex systems and make pragmatic decisions without a tightly specified roadmap.

Matching Summary

This role involves owning end-to-end research and engineering projects to improve the post-training of OpenAI's agentic models used by hundreds of millions of people.

Salary

Not specified; Not specified; Not specified

Skills & Requirements

Must-have

  • Strong ML fundamentals and LLM experience
  • Hands-on experience with RL and RLHF
  • Ability to own ambiguous problems end-to-end

Nice-to-have

  • Experience with large-scale model training systems
  • Excellent taste in model behavior across domains
  • Background in building reliable machinery for experimentation

Key Requirements

  • Strong ML fundamentals
  • Hands-on LLM and RL experience
  • End-to-end project ownership capability

Work Rights

Not specified

Tailored Resume

Cover Letter