AI Red Team Security Engineer

About this role

Ethos is hiring an AI Red Team Security Engineer to conduct offensive security testing across their life insurance platform, focusing on both traditional penetration testing and emerging AI/LLM vulnerabilities. You'll identify security flaws in web apps, cloud infrastructure, and AI systems before adversaries do, working closely with engineering teams to remediate risks.

What you'll do

• Design and execute adversarial attacks against LLM systems including prompt injection, jailbreaking, and context manipulation
• Perform full-scope penetration tests across web applications, APIs, mobile apps, and cloud environments (AWS, GCP, Azure)
• Test RAG pipelines and AI agent systems for data exfiltration, unsafe tool-use, and privilege escalation vulnerabilities
• Conduct red team exercises simulating advanced persistent threat actors using MITRE ATT&CK frameworks
• Leverage AI-augmented tools to accelerate vulnerability discovery and exploit development
• Evaluate AI safety filters and guardrails for bypass techniques and develop mitigation recommendations

What they're looking for

• Penetration testing across web, APIs, mobile, and cloud platforms
• LLM and AI/ML security vulnerability assessment
• OWASP Top 10 and advanced exploitation techniques
• Cloud security (AWS, GCP, Azure) and Kubernetes assessments
• AI-augmented attack tooling and prompt engineering
• Social engineering and red team operations
• Adversarial machine learning and model extraction attacks
• Python or similar programming languages for tool development