Senior Security Engineer, Agentic Red Team

About the role

As a Senior Security Engineer on the Agentic Red Team at Google DeepMind, you will be the primary technical executor of adversarial engagements against production-level AI models, focusing on agentic behaviors such as tool usage and reasoning chains. The team works to close the "Agentic Launch Gap" by finding architectural flaws early in the design phase and helping turn findings into reusable guardrails for Google agent developers. Candidates can choose their preferred working location from Mountain View, New York or Zrich.

Key responsibilities

• Conduct rapid security assessments of agentic services, looking for vulnerabilities unique to Generative AI such as prompt injection, tool-use escalation and autonomous lateral movement.
• Design and run attack sequences that exploit non-deterministic model behavior, agentic logic errors and data-poisoning vectors.
• Write code to turn manual vulnerability findings into automated regression testing frameworks ("Auto Red Teaming") so regressions are prevented in later model versions.
• Work directly with developers during design and build to give immediate feedback and shorten the loop between offensive findings and defensive engineering.
• Maintain and expand a library of agent-specific attack patterns and exploit primitives to set release criteria for new models.

Required profile

• 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 LLM architectures, agentic workflows (e.g. chain-of-thought reasoning) and AI vulnerability classes.
• Experience developing exploits for GenAI models (prompt injection, adversarial examples, training-data extraction).
• Preferred: consulting experience with product teams or in a startup-like environment, familiarity with AI safety benchmarks, evaluation frameworks and fuzzing, and the ability to turn complex probabilistic risks into engineering fixes.

Required skills

• Excellent coding in Python, Go or C++, with experience building security tools or automation.
• Knowledge of LLMs and agentic workflows.

What we offer

• Pay of USD 174,000 to 252,000 plus a 15% bonus target, equity and benefits.