Head of Research, Professional Intelligence, DeepMind

Job description

Minimum qualifications

PhD in Computer Science, Mathematics, Applied Statistics, Machine Learning, or a related quantitative field, or equivalent practical experience. 10 years of experience in applied NLP research and productionization, including experience with advanced prompting, and LLM fine-tuning or reinforcement learning (RL). 4 years of experience leading technical or research teams and mentoring engineers.

Preferred qualifications

Startup experience iterating fast and exploring multiple strategies to optimize performance. Exposure to production systems that rely on ML models, or experience with model deployment. Significant success with building AI software and systems for specific industries, either as research publications at top PL or ML venues, or as open source artifacts. Track record of delivering high quality solutions to large, complex software problems.

Work Independently, committed to safe, transformative AI, and excellent written and verbal communication and strong stakeholder management across all levels. Demonstrated ability to deliver clean, maintainable, production-grade code in Python, adhering to sound engineering principles. About the job Working alongside our Advanced Agentic team, the Specialized Professional Intelligence team is building domain-specific AI architecture for high-stakes industries that operate on vast, sensitive, and private datasets.

The team also contributes to the broader Gemini model training effort by building evaluations, datasets and training objectives that measure performance on complex tasks. Additionally, we work on research to see how we can improve agent performance throughout these product surfaces. We are looking for roles across the board for strong infrastructure engineers, front-end engineers, and researchers excited to achieve unprecedented levels of performance.