Senior Security Engineer, Agentic Red Team, DeepMind

Google DeepMind is hiring a Senior Security Engineer for its Agentic Red Team in Zürich. The role focuses on securing agentic AI systems through rigorous testing and vulnerability identification.

The company is looking for a Senior Security Engineer to join the Agentic Red Team at Google DeepMind. This position is dedicated to evaluating the safety of agentic AI systems, uncovering potential weaknesses, and collaborating with teams to enhance overall system resilience.

Responsibilities

• Plan and conduct red team exercises targeting agentic AI models and infrastructure.
• Detect security flaws, safety hazards, and scenarios where systems might be misused.
• Work closely with research and engineering groups to fix identified issues and strengthen security postures.
• Build automated tools and frameworks to streamline security testing processes.
• Monitor emerging threats in AI security and adversarial machine learning to keep defenses current.

Requirements

• Proven expertise in cybersecurity, including hands-on experience with penetration testing or red teaming.
• Familiarity with AI and machine learning systems, specifically regarding security and safety implications.
• Strong coding abilities, particularly in Python.
• Sharp problem-solving capabilities and a meticulous approach to detail.
• Clear communication skills to effectively report risks and suggest improvements.

About the company

Google DeepMind operates as a leading artificial intelligence research laboratory. The organization is driven by a mission to solve intelligence, advance scientific discovery, and push the boundaries of AI research. It is also dedicated to fostering a diverse and inclusive workplace environment.

• Recognized as a world-leading AI research lab.
• Focused on solving intelligence and advancing scientific discovery.
• Committed to building a diverse and inclusive workforce.