Agent Post-Training Architect for API & Power Users

OpenAI; San Francisco (CA); On-site; USD 380,000 - 500,000; Full time

Job summary

OpenAI is seeking an Agent Post-Training engineer for API and power-user teams in San Francisco. You will improve capabilities, reliability, and product fit of agentic models, shaping behavior through evals, data, and post-training interventions.

Collaborate with researchers, engineers, API/product teams, Codex, infrastructure, and safety partners to decide what behaviors matter, measure them, and plan training and launches for major model runs.

Qualifications

• Strong technical fundamentals in ML, software, systems, statistics, or applied research.
• Hands-on experience with LLMs, post-training, RL/RLHF/RLAIF, evals, graders, synthetic data, and API products.
• Experience building production ML systems and developer tools.

Responsibilities

• Design evals from real developer workflows and post-training interventions.
• Build training environments around production-like tool use and performance gaps.
• Partner with API/product teams to identify high-leverage behavior gaps.
• Own end-to-end model behavior projects from analysis to launch readiness.

Skills

ML fundamentals

Software engineering

Systems

Statistics

Applied research

Tools

LLMs

Post-training

Evals

Graders

Synthetic data

Job description

OpenAI is seeking an Agent Post-Training engineer for API and power-user teams in San Francisco. You will improve capabilities, reliability, and product fit of agentic models, shaping behavior through evals, data, and post-training interventions.

Collaborate with researchers, engineers, API/product teams, Codex, infrastructure, and safety partners to decide what behaviors matter, measure them, and plan training and launches for major model runs.