Research Scientist, Multimodal Alignment, Safety, and Fairness

Artificial Intelligence could be one of humanity’s most useful inventions. At Google DeepMind, we’re a team of scientists, engineers, machine learning experts and more, working together to advance the state of the art in artificial intelligence. We use our technologies for widespread public benefit and scientific discovery, and collaborate with others on critical challenges, ensuring safety and ethics are the highest priority.

Our team's mission is to advance the frontiers of fairness, safety, and inclusion in multimodal AI, building new capabilities into Google's flagship models and products so that they work for everyone. We approach alignment research with an ecosystem view, partnering across the development and deployment cycle and grounding our work in real-world impact on global users and communities. We are a research team with a mandate to invest in longer term bets and explore innovative approaches that deliver breakthrough improvements to models (Gemini) and products. Our research advances have directly impacted multiple versions of Gemini and Imagen models.

We are seeking a Research Scientist to help leverage agentic techniques to assess and improve model safety and alignment. You will join a team working to supercharge exploration, assessment, and steering of evolving model behavior, with a focus on subjective and creative tasks. You will tackle the underlying research questions to improve collaborative specification of alignment objectives and assessment of adherence to desired behaviors.

• Design and execute experiments to validate research hypotheses
• Research and develop novel techniques for agent-human interaction workflows
• Research and implement techniques for modeling expected user behavior, identifying and summarizing loss patterns, balancing exploration and exploitation objectives

In order to set you up for success as a Research Scientist at Google DeepMind,  we look for the following skills and experience:

• PhD in Computer Science or related field.
• Strong publication record in top machine learning conferences (e.g., NeurIPS, CVPR, ICML, ICLR, ICCV, ECCV).
• Experience in at least one modern deep learning framework (e.g. Tensorflow, JAX, PyTorch). 

In addition, some of the following would be an advantage:

• Experience prototyping with VLMs with modern prompting strategies (e.g., thinking / reasoning)
• Experience finetuning and post-training LLMs using RL
• Experience with developing agentic AI solutions to complex problems
• Excited to collaborate across orgs and disciplines to leverage diverse perspectives and expertise and find creative solutions. 
• Interest in AI alignment/fairness/safety/responsibility.

The US base salary range for this full-time position is between $141,000 - $202,000 + bonus + equity + benefits. Your recruiter can share more about the specific salary range for your targeted location during the hiring process.

At Google DeepMind, we value diversity of experience, knowledge, backgrounds and perspectives and harness these qualities to create extraordinary impact. We are committed to equal employment opportunity regardless of sex, race, religion or belief, ethnic or national origin, disability, age, citizenship, marital, domestic or civil partnership status, sexual orientation, gender identity, pregnancy, or related condition (including breastfeeding) or any other basis as protected by applicable law. If you have a disability or additional need that requires accommodation, please do not hesitate to let us know.