Skip to content
AI.info

jobs

Research Scientist, Gemini Agent Safety Post Training, DeepMind

The Gemini Safety team is accountable for the standards of GDM’s flagship releases. In this role, you will develop algorithmic solutions to advance user-facing architectures.

Company
Google
Location
Mountain View, CA, USA
Status
Closed
Posted
2026-09-24 15:40 UTC

Google DeepMind's Gemini Safety team is hiring a research scientist to work on safety and alignment for Gemini and other generative models. The role involves designing protocols that surface safety gaps in model behavior, feeding findings back into post-training, and applying supervised fine-tuning, reinforcement learning and reward modeling at scale, alongside evaluations and cross-functional work with post-training and product teams. A PhD or equivalent experience plus one year with generative AI, LLMs, NLP or agents is required; AI safety experience and publications are preferred. On-site in Mountain View, full-time.

Original job posting