AI Security Researcher & Red Team Architect

Webhosting; Austin (TX); Hybrid; USD 180,000 - 280,000; Full time

Benefits offered by this job

Equity plan

Job summary

Cloudflare in the United States, Austin and New York, seeks an AI Security Research & Red Team Engineer to advance our AI threat emulation within the Security Threat Detection, Response and Emulation org. You will drive research, exercises and attacker emulation to improve security posture for AI-driven threats.

Ideal candidates have 4+ years in offensive security, deep AI/LLM knowledge, and experience with MITRE ATT&CK; you will mentor peers and translate findings into actionable defenses.

Qualifications

• 4+ years in offensive security, application security or other relevant field.
• Deep knowledge of AI, coding agents, LLMs, prompt engineering, AI-related attack vectors (e.g., prompt injection, jailbreaking), and Agentic concepts.

Responsibilities

• AI Security and Vulnerability Research: Stay current with emerging threats and perform deep-dive research to identify AI-specific vulnerabilities and risks.
• Agentic testing and adoption: Identify AI-related attack surfaces through testing of agentic implementations and LLM usage to define requirements and guidance.
• Adversary Simulation: Conduct full-chain red team operations targeting Cloudflare’s infrastructure, networks, and ecosystems.
• Efficacy Testing: Establish a framework for testing security detections against known TTPs.
• Purple Teaming: Collaborate with the Blue Team to translate findings into defensive improvements.
• Mentorship & Growth: Lead technically while fostering curiosity and ethical hacking.
• Executive Reporting: Translate exploits into risk-based narratives for leadership.

Skills

Offensive security

Application security

AI/LLM knowledge

MITRE ATT&CK

Communication skills

BAS tooling

Tools

BAS tools

C2 frameworks

Job description

Cloudflare in the United States, Austin and New York, seeks an AI Security Research & Red Team Engineer to advance our AI threat emulation within the Security Threat Detection, Response and Emulation org. You will drive research, exercises and attacker emulation to improve security posture for AI-driven threats.

Ideal candidates have 4+ years in offensive security, deep AI/LLM knowledge, and experience with MITRE ATT&CK; you will mentor peers and translate findings into actionable defenses.