Requirements

• Experience leading teams focused on AI safety, security, or responsible AI research, including adversarial robustness, red teaming, safety evaluations, or mitigations for frontier models.
• Experience with areas such as automated red teaming, jailbreak discovery, LLM-based autoraters, evaluation design, safety classifiers or probes, prompt injection, agent security, or evaluation of severe-harm risks (e.g., CBRN or harmful manipulation).

Nice to haves

• Master’s degree or PhD in Engineering, Computer Science, or related technical field, with focus on adversarial machine learning or AI/ML security.
• Proven research or engineering achievements, such as publications in peer-reviewed venues, open-source projects, or public benchmarks.

What you'll be doing

• Leading a team of Research and Software Engineers to build engineering foundations for DeepMind’s Responsibility research portfolio.
• Developing automated platforms for discovering, measuring, and mitigating jailbreaks, red-teaming tools, and defenses for safe AI operation on the open web.
• Working on evaluations and tools for mitigations against severe harms, including chemical, biological, radiological, and nuclear (CBRN) risks and harmful manipulation.
• Adapting quickly to the evolving responsible AI landscape and directing engineering efforts where most impactful.
• Balancing research experiments with scalable infrastructure relied upon by other teams.
• Hiring, developing, and retaining a high-performing engineering team.
• Setting priorities and aligning strategy across multiple workstreams within the responsibility organization.
• Setting clear individual expectations, providing feedback and coaching, and supporting team development.
• Developing and evolving long-term technical goals and roadmaps for evaluation pipelines, red-teaming tools, autoraters, and mitigation training pipelines.
• Overseeing system designs and reviewing code to address ambiguous problems, ensuring code quality, scalability, reliability, and secure handling of sensitive data.
• Partnering with Research Scientists, Subject-Matter Experts, and various teams to translate research into evaluations and mitigations that inform model launch decisions.
• Presenting trade-offs and results to senior leadership.

Perks and benefits

• Opportunity to join a pioneering AI lab focused on advancing AI for global challenges and scientific discovery.
• Collaborative, interdisciplinary teams with diverse learning opportunities and varied career pathways.
• Commitment to safety, ethics, and widespread public benefit in AI development.