Research Engineer, Agentic AI & Tooling

OpenAI; California (MO); On-site; USD 180,000 - 300,000; Full time

Job summary

OpenAI is seeking a highly capable Codex Research team member to improve agentic models' capabilities, reliability, and product fit. You may own a research direction, build infrastructure for large-scale training, or create evals that reveal model failures, driving improvements from idea to launch.

This role supports collaboration with researchers, engineers, product teams, and safety partners to determine what enters major model runs and ship improvements to production systems.

Qualifications

• Strong fundamentals in ML, software engineering, systems, or statistics.
• 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.

Responsibilities

• Design and run experiments that improve agentic model behavior across coding, tool use, function calling, and multi‑agent tasks.
• Own end‑to‑end improvements to the post‑training stack including RL, data pipelines, graders, reward signals, evals, diagnostics, and model‑behavior analysis.

Skills

LLMs

RL

Post-training

Evaluations

Graders

Synthetic data

Tool‑using agents

Production ML systems

Machine learning fundamentals

Software engineering

Systems

Statistics

Education

Bachelor’s or higher in a technical field

Tools

Python

ML infra/tools

Job description

OpenAI is seeking a highly capable Codex Research team member to improve agentic models' capabilities, reliability, and product fit. You may own a research direction, build infrastructure for large-scale training, or create evals that reveal model failures, driving improvements from idea to launch.

This role supports collaboration with researchers, engineers, product teams, and safety partners to determine what enters major model runs and ship improvements to production systems.