The Senior Security Engineer role on the Agentic Red Team at DeepMind focuses on identifying and mitigating vulnerabilities in generative AI models. This specialized team aims to close the Agentic Launch Gap by conducting adversarial engagements and providing immediate feedback during the design phase of AI products. The engineer will work closely with product builders to ensure security measures are integrated early in the development process. This position requires a strong foundation in offensive security and a passion for addressing complex challenges in AI safety.

• Bachelor's degree in Computer Science, Information Security, or equivalent practical experience.
• 5 years of experience in Red Teaming, Offensive Security, or Adversarial Machine Learning.
• Experience with Large Language Model (LLM) architectures and AI vulnerability classes.
• Experience developing exploits for GenAI models, including prompt injection and adversarial examples.
• Excellent coding skills in Python, Go, or C++.

• Conduct rapid, high-impact security assessments on agentic services focusing on vulnerabilities unique to Generative AI.
• Engineer and execute attack sequences that exploit non-deterministic model behaviors and agentic logic errors.
• Write code to transform manual vulnerability discoveries into automated regression testing frameworks.
• Partner directly with developers during the design and build phases to provide immediate feedback.
• Maintain and expand a library of agent-specific attack patterns and exploit primitives.

• Employees at Google are often offered benefits like comprehensive health insurance, 401(k) matching, and flexible work arrangements, among other benefits.