AI Security Research & Red Team Engineer

An engineer on Cloudflare's Red Team, within its Security Threat Detection, Response and Emulation organisation, focused on research and attacks against AI, agents and LLMs as well as broader adversary simulation. The role is based in Austin or New York.

The engineer researches AI-specific vulnerabilities, tests agentic implementations and LLM use, runs full-chain red team operations against infrastructure and corporate systems, measures how well WAF, EDR and SIEM detections hold up, works with detection and incident response teams in purple exercises, and reports risk to leadership. Cloudflare wants four or more years in offensive or application security, deep knowledge of prompt injection and other AI attacks, manual pen testing or exploit development, MITRE ATT&CK, and C2 tooling familiarity.