Research Scientist, Gemini Information Tasks

We are seeking a research scientist to precisely improve Gemini's information-seeking capabilities. The successful candidate will work on post-training research in Gemini, focusing on quality of information-seeking responses. This role offers an opportunity to explore fundamental issues in modelling and data interventions for information-seeking scenarios, with significant opportunities in shaping Google's products in this space.

Responsibilities:

• Conduct research on post-training methods for information-seeking scenarios in Gemini, including reinforcement learning and self-supervised training.
• Develop novel evaluation methods for improving model quality, grounding, and factuality.
• Investigate orchestration of tool calls and improved retrieval methods for information-seeking scenarios.

Requirements:

• PhD in a relevant area, or an equivalent research/publication record.
• Strong software-engineering skills in addition to a research background.

Preferred Qualifications:

• Experience in reinforcement learning.
• Experience in post-training methods.
• Experience in Large Language Models for information-seeking scenarios.

The US base salary range for this full-time position is between $147,000 USD - 211,000 + bonus + equity + benefits.