Summary
Liam Purvis is a Senior Data Scientist in New York with 11 years of experience building production ML systems that drive revenue and operational improvements. He has a track record of shipping end-to-end solutions—market-rate forecasting that produces ~36,000 daily forecasts, a company-wide A/B testing engine, and an auction-theory-backed automation that improved margins by ~15%. At Digit and Gopuff he designed high-impact models for overdraft prevention and multi-product demand forecasting, scaling to hundreds of stores and hundreds of thousands of users. His background blends rigorous statistics (UC Berkeley) and engineering (MEng EECS), enabling him to move projects from research clubs and POCs into monitored production. Notably, he has built large-scale data infrastructure—including a 4+ billion row database on Google Cloud—and co-invented a patented ML ensemble earlier in his career. Curious about both the right question and the right answer, he combines quantitative depth with product-focused delivery.
11 years of coding experience
10 years of employment as a software developer
Bachelor's degree, Statistics, Bachelor's degree, Statistics at University of California, Berkeley
Master of Engineering - MEng, Electrical Engineering and Computer Science, Master of Engineering - MEng, Electrical Engineering and Computer Science at UC Berkeley College of Engineering
Associate of Arts - AA, Associate of Arts - AA at Ulster County Community College
De Anza College