Software Engineer, Codex Core Agents
OpenAI · San Francisco
About the Team The Codex Core Agent team builds the kernel of Codex. We own making the agent better, accelerating research, and making those improvements real in production for our users. That means working across the systems that make Codex actually function as an agent in the real world: the production performance envelope around tokens, latency, reliability, cost, and capacity; the core execution loop and interfaces that turn models into useful behavior; the shared infrastructure that enables other teams to build on Codex; and the feedback loops that turn real-world usage into better models and better agent behavior over time. About the Role We’re looking for engineers to build the infrastructure that powers Codex agents in production. This role focuses on the systems that let models safely execute code, interact with tools, complete long-running tasks, and operate reliably and efficiently at scale. You’ll design and operate the infrastructure behind sandboxed execution, orchestration, stateful workflows, app-server and SDK boundaries, and model rollouts. You’ll work at the intersection of distributed systems, developer tooling, and AI, building primitives that make Codex faster, safer, more reliable, and easier for the rest of the organization to build on. What You’ll Do - Design and build execution environments for AI agents, including sandboxing, isolation, and reproducibility. - Develop systems for agent orchestration across multi-step, tool-using workflows. - Build infrastructure for running, testing, and debugging code generated by models. - Create state and memory systems that allow agents to persist context across long-running tasks. - Optimize tokens, latency, reliability, and cost across Codex’s production fleet. - Support model rollouts, capacity planning, and the core tradeoffs between quality, speed, and economics to manage a fleet of frontier agents at scale. - Build shared platform capabilities that unblock product teams, part