The Research Scientist role at DeepMind focuses on enhancing the safety and behavior of AI systems, particularly within the Gemini project. The team is dedicated to developing innovative auditing frameworks and defensive safeguards for machine intelligence. This position requires a strong technical contributor who can take research from concept to product while ensuring ethical integrity and alignment standards. The role emphasizes collaboration and rapid incident mitigation in a fast-paced environment.

• PhD degree in Computer Science, a related field, or equivalent practical experience.
• 2 years of experience in Large Language Model safety or security.
• Experience in developing and leveraging agentic workflows around safety, behavior, and alignment.
• Experience with synthetic data generation pipelines and building evaluations for non-verifiable tasks.
• Strong experimental taste with good judgment regarding baselines and testing.

• Drive innovation and understanding of safety and security jailbreaks, delivering scalable mitigations.
• Develop red and blue teaming methods for frontier GenAI models, providing actionable insights.
• Explore data and algorithmic solutions to ensure Gemini Models are safe and helpful.
• Improve Gemini’s adversarial capabilities focusing on high-stakes abuse risks.
• Develop and execute experimental plans to address known gaps or create new capabilities.

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