jobs

Agent Post-Training, Artifacts Research

OpenAI’s Agent Post-Training team is seeking a researcher or engineer to train frontier models to create polished, useful work products such as documents, spreadsheets, slide decks, dashboards, reports, analyses, and interactive artifacts.

In inglese

Company

OpenAI

Location

San Francisco, United States

Status

Closed

Posted

2026-09-15T03:33:35.206+00:00

OpenAI's Agent Post-Training team is hiring a researcher or engineer to train frontier models to produce work artifacts: documents, spreadsheets, slide decks, dashboards, reports and interactive outputs built from vague goals. The work covers the post-training stack — reinforcement learning, data pipelines, graders, reward signals, evaluations and diagnostics — plus environments that expose model failures. Synthetic data, early-training and alignment interventions, latency and launch readiness feature, alongside partnership with the Codex and ChatGPT teams. The posting asks for ML or software-engineering fundamentals and hands-on LLM, RL, RLHF/RLAIF or production ML experience.