Context Architect for Frontier Agent Training

Neura Market; San Francisco (CA); On-site; USD 150,000 - 190,000; Full time

Job summary

OpenAI is seeking a Context Researcher on the Agent Post-Training team to scale compute spent on context and to drive improvements in the post-training stack. You will work with researchers, engineers, product, infrastructure, and safety partners to decide what goes into major model runs and ship enhancements to frontier products used by developers and enterprises.

You will design experiments, build evals and data pipelines, and contribute to multi-agent and production ML systems, ensuring

Qualifications

• Strong technical foundations in ML, software engineering, systems, or statistics.
• Hands-on experience with LLMs, RL, RLHF/RLAIF, post-training and evals.

Responsibilities

• Design experiments to scale compute on context for frontier models.
• Own end-to-end post-training improvements across data, graders, signals and diagnostics.
• Build evals/environments to surface model failures and guide training data or research.
• Collaborate with Codex/ChatGPT product teams to translate signals into improvements.
• Work on early-training/alignment interventions and integration readiness.

Skills

ML fundamentals

Software engineering

Systems

Statistics

LLMs

RL / RLHF / RLAIF

Post-training

Evals

Synthetic data

Production ML systems

Job description

OpenAI is seeking a Context Researcher on the Agent Post-Training team to scale compute spent on context and to drive improvements in the post-training stack. You will work with researchers, engineers, product, infrastructure, and safety partners to decide what goes into major model runs and ship enhancements to frontier products used by developers and enterprises.

You will design experiments, build evals and data pipelines, and contribute to multi-agent and production ML systems, ensuring