Research Scientist, AI Safety and Alignment, DeepMind
About The Position
The Research Scientist will conduct research focused on AI safety and alignment within Google DeepMind’s Deep Understanding team, which studies machine behavior, capabilities, and limitations to help safely empower people. The team combines empirical science with practical engineering, prioritizes concrete safety improvements, and works on both long-term safety questions and vulnerabilities in live systems. The role involves setting up large-scale tests, rapidly deploying promising ideas, managing research deliverables, applying current theories to develop technologies and products, and contributing findings to the wider research community through publications and collaboration. The position supports research across Google’s portfolio, spanning fundamental research, product innovation, product contribution, and infrastructure goals.
Requirements
- PhD in Computer Science, Machine Learning, Artificial Intelligence, Mathematics, Physics, or a related STEM discipline, or equivalent practical experience.
- At least 1 year of experience initiating, scoping, and executing research in AI safety, alignment, or mechanistic interpretability.
- Experience evaluating, steering, red teaming, or implementing empirical safeguards on Large Language Models (LLMs) or frontier systems.
- Experience with Python or C++, deep learning frameworks such as JAX or PyTorch, and agentic workflows or tools.
- First-author publications at machine learning conferences such as NeurIPS, ICML, ICLR, or COLM.
Nice To Haves
- Post-doctoral or industry experience driving prototype-driven safety projects, with recognized publications, seminal papers, or open-source research frameworks.
- Interdisciplinary research experience bridging cognitive science, computer science, machine learning, and technological ethics to address systemic AI risks.
- Deep expertise in post-training alignment, reinforcement learning, reasoning models, and agentic or mechanistic interpretability to diagnose machine behaviors.
- A demonstrated track record of translating theoretical alignment and safety findings into scalable, production-grade mitigations on live frontier models.
- Exceptional technical communication skills and the ability to influence cross-functional engineering, product, and safety policy stakeholders.
Responsibilities
- Define data structures, frameworks, research designs, and evaluation metrics for research solution development and implementation.
- Identify project timelines and obtain needed resources.
- Identify problems or gaps in existing technology and engage stakeholders and leaders to address them.
- Author research papers and share research results across functions and with the research community.
- Identify emerging research areas through interaction with internal and external collaborators.
- Help develop long-term research strategy and plans to expand the impact of Google research.
- Share research trends and best practices within the community to help grow the research business.
Benefits
- 15% bonus target
- Equity
- Benefits are mentioned, but specific benefits are not listed.
Stand Out From the Crowd
Upload your resume and get instant feedback on how well it matches this job.
What This Job Offers
Job Type
Full-time
Career Level
Senior
Education Level
Ph.D. or professional degree
