Software Engineer, Guardian

The role

You will design and build production systems that evaluate agent behavior, identify risk, and take appropriate action. You will work across architecture, implementation, evaluation, and operations to build safety and oversight systems for AI agents.

What you will do

• Design, build, and operate core components of Auto-review and the broader Guardian platform.
• Build production systems that evaluate agent actions in real time and detect harmful behavior.
• Improve how oversight systems understand user intent, environmental context, and permissions.
• Partner with alignment researchers to turn prototypes into scalable product capabilities.
• Work across Product, Security, Infrastructure, and other engineering teams to integrate Guardian capabilities.
• Develop evaluations, adversarial tests, and monitoring to understand safety performance.
• Own projects end to end and contribute to technical direction and design reviews.
• Investigate failures to improve system behavior and reliability.

Required qualifications

• Experience building and shipping complex production software with ownership across design, implementation, deployment, and operation.
• Strong software engineering fundamentals.
• Ability to design reliable backend services, APIs, and infrastructure.
• Ability to navigate tradeoffs across safety, usability, speed, and reliability.
• Ability to translate open-ended ideas into practical systems.
• Comfort with ambiguity and ability to take initiative.
• Clear communication across technical disciplines.

Nice to have

• Built with AI agents or large language models in a core product, startup, or technically serious independent project.

About the company

OpenAI is an AI research and deployment company dedicated to ensuring that general-purpose artificial intelligence benefits all of humanity.