Description:

Seeking a Staff - Principal level offensive security domain expert to build agents for identifying and remediating vulnerabilities across OpenAI’s infrastructure and applications.
Role involves creating a production system that operates safely and reliably at scale, leveraging automation and advanced model capabilities.

Requirements:

Substantial hands-on offensive security experience with strong judgment on vulnerabilities and attack paths.
Expertise in cloud security, Kubernetes, web application security, and experience in assessing complex environments.
Ability to decompose ambiguous security problems into reliable systems and encode workflows into software.
Experience in building production-quality software and agent systems with models, tools, and feedback loops.
Strong understanding of agent reliability, evaluations, observability, and safety.

Opportunity to serve as a technical owner of an ambitious new system and make foundational architectural decisions.
Work closely with frontier models and contribute to the future of offensive security at OpenAI.

Skills