OpenAI
AI Systems Engineer, Codex Agents
$230K–$385K
San Francisco · Full-time · Posted May 15, 2026
About the job
Build and improve the Codex agent systems that safely execute tasks in production, from the agent harness and runtime infrastructure to evaluation and reliability tools.
What you'll do
- Design the agent harness and execution loop for tool use, code execution, and long-horizon tasks.
- Build sandboxing, orchestration, evaluation, experimentation, and debugging systems.
- Run experiments and improve agent reliability, solve rate, latency, cost, and production observability.
What they ask for
- Experience building or operating production systems in areas such as distributed systems, infrastructure, developer tooling, sandboxing, cloud platforms, or ML systems.
- Hands-on experience with LLM applications, coding agents, evaluations, model deployment, inference, runtime performance, or developer platforms.
- Ability to debug ambiguous production failures and deliver durable fixes.
- Strong ownership and meaningful coding; able to lead scoped or multi-team AI systems work.
Nice to have
- Deep Rust, systems, sandboxing, isolation, or low-level platform experience.
- Experience with agent harnesses, model evaluations, GPU systems, performance engineering, or open-source infrastructure.
Pay
$230K – $385K USD base salary
Tools
Rust, Python
AI focus
Build production Codex agent systems; experiment across prompts, models, and harnesses, and work with research to make the harness useful for training and evaluating agentic models.
Model-written summary · the original posting is authoritative