Zurich, Switzerland
DeepMind is hiring a Research Scientist to develop post-training recipes for frontier models, especially related to Gemini. The role involves architecting Reward Modeling and Reinforcement Learning strategies and advancing capabilities. Candidates need a PhD in a related field and experience with LLMs and RL.
Large Language Models (LLMs), Reinforcement Learning (RL), Reward Modeling, experiment design, coding
Not specified