Research Scientist, Gemini Safety
Mountain View, California, US
Posted 3mo ago
Job Location
Mountain View, California, US
Tech Stack
Remote Work Policy
On-site
Categories
AI Research Engineer
About the job
The Gemini Safety team at Google DeepMind is responsible for the safety and fairness of the latest Gemini models. As a Research Scientist/Engineer, you will apply and develop cutting-edge data and algorithmic solutions to advance these user-facing models. This is a fast-paced, highly collaborative role within a supportive team dedicated to pushing the boundaries of AI for public benefit and scientific discovery, with safety and ethics as the highest priorities.
Responsibilities
- Post-training and instruction tuning of LLMs, focusing on text-to-text, image/video/audio-to-text modalities, and agentic capabilities.
- Exploring data, reasoning, and algorithmic solutions to ensure Gemini Models are safe, helpful, and equitable.
- Improving Gemini's adversarial robustness, particularly for high-stakes abuse risks.
- Designing and maintaining high-quality evaluation protocols to assess model behavior gaps and headroom related to safety and fairness.
- Developing and executing experimental plans to address known gaps or build new capabilities.
- Driving innovation and enhancing understanding of Supervised Fine Tuning and Reinforcement Learning fine-tuning at scale.
Requirements
- PhD in Computer Science or a related field, or equivalent practical experience.
- Significant LLM post-training experience.