Principal Security Researcher

A principal researcher on GitLab's Application Security team, reporting to the Senior Manager of Application Security, who attacks the company's AI features: the Duo Agent Platform, Duo Chat and agentic workflows, as well as the wider codebase and the open source components it relies on.

The researcher hunts for systemic and chained flaws, proves them with working exploits, devises new ways to test AI agents, sets security requirements for engineering teams, builds agent-assisted discovery tooling and mentors others. GitLab wants a decade or more of vulnerability research, pentesting or other offensive work, fluency in two or more of Ruby, Go, Python, TypeScript and Rust, and a firm grasp of prompt injection and agent manipulation. Remote in the US, Canada, the UK or Israel.