Research Scientist, Model Collaborativity

Snapshot

Artificial Intelligence could be one of humanity’s most useful inventions. At 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.

About Us

We aim to enhance Gemini's capabilities to become a truly collaborative partner. User intent is typically unveiled through an ongoing, interactive, and multi-step conversation, rather than a single, isolated prompt.

We seek for the model to be able to proactively engage by asking clarifying questions, making timely suggestions, and automatically incorporating relevant context from past interactions and personal data.

This challenge is conceptualized as an imperfect information game. The core objective is for the model to deduce a user's hidden goals and preferences through cues provided during a conversation. We plan to utilize advanced Reinforcement Learning methods to optimize for long-term user satisfaction, which necessitates solving intricate credit assignment issues within interactive, stateful environments; model robustness issues; and exploration/exploitation strategies.

The techniques you develop will have a direct impact on a wide range of Gemini applications.

Key responsibilities:

• Design and implement novel multiturn RL algorithms to train collaborative LLMs. This includes exploring advanced methods for credit assignment, model robustness, and exploration/exploitation strategies.
• Develop and scale our training infrastructure, building on our existing framework for training against stateful user simulators.
• Formalize the problem of collaborativity by creating new metrics, environments, and evaluation methodologies that capture long-term user satisfaction and preference alignment.
• Do cutting-edge research that pushes the boundaries of how agents learn from interactions with users, user simulators and other agents with diverse model behaviors.
• Collaborate with research and product teams to integrate these capabilities into core Gemini products, improving tasks that require sustained interaction and user understanding.

To make this effort successful, we need a strong RS who can help us deliver state-of-the-art collaborative models. We are looking for a candidate with deep expertise in reinforcement learning and large-scale ML systems. You should be passionate about solving complex, long-horizon problems and excited by the challenge of building truly adaptive and intelligent agents.

About You

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

• PhD in Machine Learning, Reinforcement Learning, Natural Language Processing, or a related field.
• Strong data analysis and synthetic data generation skills.
• Strong development skills in Python and experience with deep learning frameworks like JAX, PyTorch, or TensorFlow.
• Experience building and working with large-scale ML training systems.

In addition, the following would be an advantage: 

• Deep theoretical and practical experience in Reinforcement Learning (e.g., policy gradient methods, value-based methods, model-based RL, credit assignment, robustness).
• Experience developing and training large generative models (LLMs).
• Strong track record of academic publications in top-tier conferences (e.g., NeurIPS, ICML, ICLR, AAAI).
• Familiarity with research on game theory, multi-agent systems, or learning from human feedback (RLHF/RLAIF).
• Experience building or using user simulators for RL training.

The US base salary range for this full-time position is between $166,000 - $291,000 + bonus + equity + benefits. Your recruiter can share more about the specific salary range for your targeted location during the hiring process.

Application Deadline: December 15, 2025

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.