Summary
Zifan Lin is a data scientist with 11 years of experience applying statistical rigor to AI and risk problems, currently driving AI response quality for Google Search in New York. Prior roles at Google DeepMind and the Gemini team focused on LLM evaluation, factuality assessment, and A/B "arena" experiments to shape model releases, reflecting deep expertise in evaluation methodology and user-signal analysis. Earlier work in credit risk modeling and model validation at State Street and Huntington honed robust practices for model reliability and regulatory contexts. Zifan blends hands-on experimentation with a detail-oriented, self-driven approach and a Columbia MA in Statistics, enabling both product-facing insights and defensible metrics. Notably, they bridge enterprise risk discipline with cutting-edge LLM evaluation, making them adept at translating complex model behavior into actionable quality improvements.
11 years of coding experience
8 years of employment as a software developer
High School Diploma, High School Diploma at Hangzhou Foreign Languages School
Bachelor's degree Statistics, Bachelor's degree Statistics at Renmin University of China
Master of Arts (MA) Statistics, Master of Arts (MA) Statistics at Columbia University
Chinese, English