Wendy is an ML Engineer specializing in search, recommendation, and personalization systems at scale, with 10 years of experience spanning social networks (Twitter/X, Pinterest, Meta), e-commerce (Coupang), and marketplace platforms. She currently works at Airbnb building search and recommendation systems.
Her academic background includes a PhD in Statistics with a minor in Computer Science from The Ohio State University, with doctoral research in network sampling and estimation theory. She founded TigerLab AI, an open-source toolkit for LLM safety evaluation, and has taught online courses on Python and machine learning.
Wendy is the author of Advanced Retrieval-Augmented Generation: Bridging Large Language Models and Knowledge Graphs (Wiley-IEEE Press) and Data Science: Methods and Practice (China Machine Press) — two technical books bridging research and industry practice. Beyond technical work, she is committed to teaching, mentoring, and advancing best practices for scalable, production-grade ML systems.
A complete guide from the foundations of information retrieval to the cutting-edge frontiers of RAG. Bridging LLMs and knowledge graphs, this book provides theoretical principles, practical techniques, and hands-on frameworks for building reliable AI systems that minimize hallucinations and improve factual correctness. Covers Graph-RAG architectures, KG construction, RAG pipeline engineering, and production-ready implementations using LlamaIndex, Neo4j, and leading Graph-RAG frameworks.
A comprehensive data science textbook spanning 5 parts and 16 chapters — covering statistics, machine learning, deep learning, data engineering, product analytics, A/B experimentation, and domain applications in search, recommendation, advertising, NLP, and large language models. Combines methods with real-world industry practice.
Dissertation: On Estimation Problems in Network Sampling
Research: Statistical methods for estimation challenges arising from sampled network data structures
Award: Winner, 2013 Capital One National Data Analytics/Modeling Competition
Methods and systems for identifying automated, scripted, or malicious interactions on social media platforms using machine learning-based detection models.