Improving Gemini for information tasks, focusing on the quality of information-seeking responses (helpfulness, factuality, grounding, and other aspects). This involves exploring fundamental issues in modeling and data interventions for information-seeking scenarios to shape Google's products.
Requirements
- Strong software-engineering skills in addition to a research background
- Experience in reinforcement learning
- Experience in post-training methods
- Experience in LLMs for information-seeking scenarios
Responsibilities
- Research on post-training (e.g., RL and SFT) for information-seeking scenarios in Gemini
- Research on novel evaluation methods for improving model quality, grounding and factuality
- Research on orchestration of tool calls, and improved retrieval methods, for information-seeking scenarios
Other
- PhD in a relevant area, or an equivalent research/publication record
- Number of years experience: anything from recent PhD onwards
- The US base salary range for this full-time position is between $141,000 - $202,000 + bonus + equity + benefits.
