Applied AI Engineer, Codex Agents & Production Reliability

Neura Market; San Francisco, Northern (CA, KY); Hybrid; USD 180,000 - 250,000; Full time

Job summary

OpenAI is seeking applied AI engineers to advance Codex agents from impressive demos to dependable tools. You will improve agent performance on real software tasks and translate research gains into practical, measurable improvements for users.

You’ll collaborate with researchers, infrastructure, and product to ensure agents are powerful, steerable, and reliable in practice, delivering meaningful gains in solve rate and economic value for customers.

Qualifications

• Experience building or shipping machine learning or LLM-powered products.
• Strong in Python and comfortable with modern ML tooling.
• Experience with model evaluation, fine-tuning, or prompt design.
• Think in terms of systems and user outcomes, not just model metrics.
• Enjoy debugging messy, real-world failures and turning them into improvements.

Responsibilities

• Design and iterate on agent behaviors across real-world coding tasks and long-horizon workflows.
• Work closely with research to develop and run evals to measure agent performance, regressions, failure modes, and edge cases.
• Improve performance through prompting, tool-use strategies, context construction, and model-facing experimentation.
• Analyze failures in production and systematically improve robustness and reliability.
• Build feedback loops and data systems that get better real-task data into evaluation and research.
• Work with product teams to shape user-facing agent experiences and the interfaces the agent depends on.
• Help define what “good” looks like for agents completing complex tasks end-to-end.

Skills

Python

ML tooling

Model evaluation

Fine-tuning

Prompt design

Systems thinking

Debugging

Tools

Agent frameworks

Tool-using LLM systems

Production logs analysis

Job description

OpenAI is seeking applied AI engineers to advance Codex agents from impressive demos to dependable tools. You will improve agent performance on real software tasks and translate research gains into practical, measurable improvements for users.

You’ll collaborate with researchers, infrastructure, and product to ensure agents are powerful, steerable, and reliable in practice, delivering meaningful gains in solve rate and economic value for customers.