jobs
Research Scientist, User Signals from Agentic Logs, DeepMind
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…
In inglese
- Company
- Location
- Zürich, Switzerland
- Status
- Open
- Posted
- 2026-10-20 14:38 UTC
A research scientist on Google DeepMind's Gemini User Signals team, based in Zürich, working on post-training: turning user interactions and feedback into training signals. The work covers reward and critique models for RLHF, preference optimization, data selection and weighting, and evaluation frameworks that link offline reward-model accuracy to online user satisfaction, with recipes meant to carry across model sizes and text and audio modalities. The posting asks for a PhD or equivalent, Python with PyTorch or JAX, and published research in post-training, RL or reward modeling.
Original job posting