Security Engineer, Agent Security

About this role

OpenAI is hiring a Security Engineer to design and implement security frameworks, policies, and controls for agentic AI systems. You'll develop threat models, build security tooling, collaborate across teams, and help shape industry standards for securing autonomous AI.

What you'll do

• Design and implement identity, network, and runtime-level security controls (sandboxing, policy enforcement) integrated with Agent Infrastructure
• Develop and ship production-grade security tooling to harden safety monitoring pipelines at scale
• Collaborate with Agent Infrastructure, product, research, safety, and security teams to balance security, performance, and usability
• Architect isolation strategies and defenses for emerging agentic AI technologies
• Shape the Agent Security roadmap and help define industry standards for securing autonomous AI
• Influence security strategy through clear communication with engineering, research, and leadership

What they're looking for

• Software engineering in Python, Go, Rust, or C/C++
• Container security and kernel-level hardening expertise
• Modern isolation techniques and sandbox technologies
• Network security and identity-based access controls
• Cloud security (Azure, AWS, or GCP) including IAM and infrastructure-as-code
• Experience building and operating high-reliability, secure services
• Cross-functional communication and influence skills
• AI/ML security knowledge (nice-to-have)