Summary
Silvester Yao is a machine learning software engineer with 11 years of experience building large-scale data and ML systems across Meta, LinkedIn, and Experian, currently improving Feed relevance and PYMK at Meta using SOTA content/user understanding and LLMs. Trained in applied statistics at Purdue, he combines rigorous statistical modeling with production engineering—designing TB-scale Spark pipelines, real-time data health monitors, and feature-generation DSLs that scaled feature engineering 100x. He moves fluidly between research-grade models (NLP, vision, graph embeddings, MTML) and engineering constraints, having delivered measurable DAU growth and robust data-quality platforms. Based in Sunnyvale, he pairs an actuarial attention to risk with a programmer’s drive for efficiency, and is known for turning complex compliance and risk datasets into actionable, production-ready features.
11 years of coding experience
7 years of employment as a software developer
Master’s Degree Applied Statistics, Master’s Degree Applied Statistics at Purdue University
Chinese, English