Senior Data Scientist, Research, Generative AI Factuality

Senior Data Scientist, Research, Generative AI Factuality

In this role, you will help evaluate and improve the quality of tools and products that apply AI to improve access to accurate and helpful information.

The team is responsible for Gemini factuality and quality on info-seeking prompts. The goal is to ensure that models are as accurate as possible without compromising on any other response quality aspect, and ensuring that response structure and organization are helpful. The team is also responsible for factuality evaluations in Gemini, including defining factuality metrics, eval templates, and building high-quality autoraters. This role partners with various Gemini verticals: Factuality, Quality, Multi-Turn, Multilinguality, and Model Behavior.

Responsibilities include collaborating with stakeholders in cross-projects and team settings to identify and clarify business or product questions to answer, and provide feedback to translate and refine business questions into tractable analysis, evaluation metrics, or mathematical models. Use custom data infrastructure or existing data models as appropriate, and design and evaluate models to mathematically express and solve defined problems with limited precedent. Gather information, business goals, priorities, and organizational context around the questions to answer, as well as the existing and upcoming data infrastructure. Own the process of gathering, extracting, and compiling data across sources via relevant tools (e.g., SQL, R, Python). Format, re-structure, or validate data to ensure quality, and review the dataset to ensure it is ready for analysis.

Minimum qualifications include a Masters degree in Statistics, Data Science, Mathematics, Physics, Economics, Operations Research, Engineering, a related quantitative field, or equivalent practical experience. Five years of work experience using analytics to solve product or business problems, coding (e.g., Python, R, SQL), querying databases or statistical analysis, or three years of work experience with a PhD degree. Preferred qualifications include eight years of work experience using analytics to solve product or business problems, coding (e.g., Python, R, SQL), querying databases or statistical analysis, or six years of work experience with a PhD degree. Experience with LLM evaluation, including autoraters and log analysis.