Research Scientist, Mechanistic Interpretability

About the job

At DeepMind, we've built a unique culture and work environment where long-term ambitious research can flourish. We are seeking a highly motivated Research Scientist to join our team and contribute to groundbreaking research at the intersection of mechanistic interpretability, cognitive science, and the philosophy of mind. You will investigate cognitive processes in large language models, helping us understand the inner workings of Gemini and other models and advancing the path toward safe, aligned Artificial General Intelligence (AGI).

Artificial intelligence will be one of humanity’s most transformative inventions. At Google DeepMind, we are a pioneering AI lab with exceptional interdisciplinary teams focused on advancing AI development to solve complex global challenges and accelerate high-quality product innovation for billions of users. We use our technologies for widespread public benefit and scientific discovery, ensuring safety and ethics are always our highest priority.

We are pushing the boundaries across multiple domains. Our global teams offer diverse learning opportunities and varied career pathways for those driven to achieve exceptional results through collective effort.

Responsibilities

• Define and lead a novel research agenda focused on the mechanistic interpretability of Large Language Model (LLMs) including self-conceptions, theory of mind and social media platform-cognition.
• Adapt, and extend state-of-the-art mechanistic interpretability techniques from the broader research community, applying them directly to Gemini and other models.
• Design and execute experiments to identify internal vectors and mechanisms that influence a model's outputs, especially where they appear reflective or pertain to its conception of itself and of other models.
• Collaborate cross-functionally with technical alignment teams, bridging interpretability research with broader AI safety and capabilities initiatives.
• Publish and communicate research findings through peer-reviewed papers, internal technical reports, and presentations to academic and cross-disciplinary audiences.

Qualifications

Minimum qualifications:

• PhD in Computer Science, Mathematics, Physics or a related technical field, or equivalent research experience.
• 3 years of experience designing, executing, and analyzing empirical research in AI.
• Experience using machine learning frameworks (e.g., PyTorch, JAX) to run experiments.
• Experience working with large language model architectures and their internal representations.
• Experience communicating scientific concepts to technical and non-technical audiences.
• Publication record in mechanistic interpretability, AI safety, or machine learning in peer-reviewed venues.

Preferred qualifications:

• Postdoctoral or equivalent industry research experience in mechanistic interpretability.
• Experience or demonstrated interest working across the cognitive sciences, philosophy of mind, psychology, or neuroscience.
• Experience evaluating concepts of selfhood, metacognition, and theory of mind in artificial systems.
• Experience translating foundational research into organizational impact, benchmarks, or open-source tools.
• Participation in, or mentorship of, specialized AI safety and interpretability research programs.
• Track record of interdisciplinary collaboration, especially bridging technical machine learning with philosophical concepts.