Applied AI Engineer, Codex Core Agent

Sourced from simplify.jobs

Applied AI engineer on OpenAI's Codex Core Agent team, focused on turning Codex agents from demos into dependable tools for real software engineering tasks. The role spans agent behavior design, evaluation, prompting, tool-use strategies, and production reliability. You'll work closely with research, infrastructure, and product to convert model improvements into measurable gains in solve rate and user value.

• Design and iterate on agent behaviors across real-world coding tasks and long-horizon workflows
• 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 production failures and systematically improve robustness and reliability
• Build feedback loops and data systems that feed real-task data into evaluation and research
• Work with product teams to shape user-facing agent experiences and interfaces
• Help define what good looks like for agents completing complex tasks end-to-end

• Experience building or shipping machine learning or LLM-powered products
• Strong Python skills and comfort with modern ML tooling
• Experience with model evaluation, fine-tuning, or prompt design
• Systems-level thinking oriented toward user outcomes
• Comfort debugging messy, real-world failures and turning them into improvements

• Experience with agent frameworks or tool-using LLM systems
• Research experience with code generation models or developer tooling
• Experience working with large, messy datasets or production logs