Research Engineer / Scientist – Frontier Red Team (Cyber)

About the role

Anthropic is building a Frontier Red Team focused on AI‑enabled cyber threats. As a Research Engineer/Scientist you will design and implement tools that protect the world from advanced, autonomous AI capabilities in cybersecurity. The position is based in San Francisco and requires on‑site collaboration with security, policy, and safeguards teams.

Key responsibilities

• Develop systems, tools, and frameworks for AI‑empowered cybersecurity, including autonomous vulnerability discovery, malware detection, network hardening, and penetration testing.
• Design and execute experiments that evaluate AI cyber capabilities in realistic environments.
• Build infrastructure to enable AI models to operate safely within security contexts.
• Translate technical findings into demonstrations and artifacts for policymakers and the public.
• Collaborate with external experts in cybersecurity, national security, and AI safety to validate research directions.
• For senior candidates: define research strategy, prioritize problems, and lead the cyber‑defense research program.

Required profile

• Strong research background in AI, machine learning, or cybersecurity.
• Experience publishing high‑impact research and building prototype systems.
• Ability to work cross‑functionally with security, policy, and engineering teams.
• Demonstrated interest in AI safety and emerging cyber threats.
• Senior candidates should have a track record of leading research initiatives.

Required skills