Research Scientist, User Signals from Agentic Logs, DeepMind

Google LLC; Zürich; On-site; CHF 150,000 - 210,000; Full time

Job summary

Google DeepMind in Zürich is seeking a Research Scientist to advance the post-training strategy for our Gemini User Signals team. You will convert complex user interactions into training signals to improve model behavior and alignment.

You will lead research into reward modeling, preference optimization, and data-centric strategies, developing scalable recipes across modalities and model families. We value publications, strong Python and DL framework skills, and collaboration with

Qualifications

• PhD in ML/NLP or equivalent experience.
• Strong Python and deep learning framework skills.
• Research in post-training, RL, or reward modeling.
• First-author publications in ML/NLP venues.

Responsibilities

• Harvest and utilize high-quality user signals to improve Gemini models.
• Develop reward models and critique models for RLHF and alignment.
• Create horizontal post-training recipes across models and modalities.
• Investigate data weighting and selection to maximize signal impact.
• Establish evaluation frameworks linking offline metrics to user satisfaction.

Skills

Python

RL research

Reward modeling

Preference learning

Deep learning

Education

PhD in ML/NLP

Tools

PyTorch

JAX

Job description

Research Scientist, User Signals from Agentic Logs, DeepMind

• PhD in Machine Learning, natural language processing, or a related field (or equivalent practical experience).
• Experience with software development using Python and one or more deep learning frameworks (e.g., PyTorch or JAX).
• Research experience in post-training, reinforcement learning (RL), agentic RL, preference learning, or reward modeling.
• First-author (or co-author) publication in one or more peer-reviewed machine learning or natural language processing (NLP) conferences (e.g., NeurIPS, ICML, ICLR, ACL, EMNLP).

Preferred qualifications

• First-author publications at top ML/NLP venues (NeurIPS, ICML, ICLR, ACL, EMNLP).
• Experience with modern alignment and post-training methods (e.g., RLHF, DPO, PPO, rejection sampling, or self-play).
• Experience modeling multi-turn interaction data, agent trajectories/logs, or user feedback (implicit and explicit signals).
• Experience training or evaluating reward models, critique models, or automated evaluators (e.g., LLM-as-a-judge).
• Experience in agentic post-training.

About the job

As a part of the Gemini User Signals team, a Research Scientist will help drive the post-training strategy of our most advanced models, dedicated to establishing the "flywheel" of alignment, converting complex user interactions and feedback into training signals that fundamentally improve model behavior. In this role, you will lead research into reward modeling, preference optimization, and data-centric strategies to close the loop between user deployment and model training. You will help define the horizontal recipes that allow us to scale these insights across different modalities and model families. 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 varied learning opportunities and career pathways for those driven to achieve exceptional results through collective effort.

Responsibilities

• Design and implement mechanisms to harvest, filter, and utilize high-quality user signals (implicit and explicit) to iteratively improve Gemini models.
• Develop and refine reward models and critique models that accurately capture user intent and satisfaction, serving as the foundation for Reinforcement Learning from Human Feedback (RLHF) and other alignment techniques.
• Create and validate "horizontal" post-training recipes that can be robustly applied across different model sizes and modalities (including text and audio).
• Investigate novel methods for weighting, synthesizing, and selecting data to maximize the impact of user signals on model performance.
• Establish rigorous evaluation frameworks to measure the correlation between offline metrics (reward model accuracy) and online user satisfaction.

Google is proud to be an equal opportunity and affirmative action employer. We are committed to building a workforce that is representative of the users we serve, creating a culture of belonging, and providing an equal employment opportunity regardless of race, creed, color, religion, gender, sexual orientation, gender identity/expression, national origin, disability, age, genetic information, veteran status, marital status, pregnancy or related condition (including breastfeeding), expecting or parents-to-be, criminal histories consistent with legal requirements, or any other basis protected by law. See also Google's EEO Policy , Know your rights: workplace discrimination is illegal , Belonging at Google , and How we hire .

Google is a global company and, in order to facilitate efficient collaboration and communication globally, English proficiency is a requirement for all roles unless stated otherwise in the job posting.

Equity compensation

• Equity is granted exclusively and discretionarily by Alphabet Inc. on the basis of an agreement concluded between you and Alphabet Inc. Alphabet Inc. is your sole contractual partner with respect to equity grants. GSU grants are not guaranteed, are discretionary, are subject to approval by the Alphabet Inc. board of directors or its delegate, the terms of the relevant Alphabet Inc. stock plan, and your grant agreement. They have no impact on statutory payments. Current or past grants do not confer an acquired right.