Offensive Security Agent Engineer
Core
Build systems, harnesses, and execution environments that enable autonomous AI agents to discover and exploit vulnerabilities in modern web applications and APIs.
Role type
Senior IC Offensive Security Agent Engineer
Builds
Autonomous offensive security workflows, agent harnesses, and execution environments for the Apex and Pensar platforms
Domain
Cybersecurity, Web Application Security, Agentic AI Systems
Deliverable
production ML models | product features
Required skills
Python, Go, JavaScript/TypeScript, Rust, C/C++, browser automation (Playwright, Puppeteer, CDP), HTTP, web vulnerability classes, agentic systems, LLM applications, complex tool-using AI systems, Linux, containers, cloud infrastructure
Preferred skills
Debugging complex distributed systems, translating manual security techniques into software primitives, designing evaluation benchmarks for autonomous agents, open source contribution
Technologies
Playwright, Puppeteer, Chrome DevTools Protocol, Apex, Pensar
Responsibilities
Design and build agent harnesses for autonomous offensive security workflows; Develop tooling for agents to interact with browsers, terminals, proxies, and scanners; Build orchestration for long-running multi-step security tasks; Create evaluation systems and benchmarks to measure agent performance; Convert human offensive security techniques into reproducible agent capabilities; Contribute to the Apex open-source tool and platform architecture
Seniority
Senior, hands-on IC