I am a senior staff research engineer at Meta, where I lead the development of foundational methodologies that power experimentation and causal inference at scale. My research sits at the intersection of econometrics, causal inference, and machine learning, with a focus on building rigorous, scalable solutions for some of the most complex measurement challenges in the technology industry. 

Over the course of my years at Meta, I have pioneered industry-leading advances in online experimentation systems, tackling problems that are both technically deep and practically impactful. My work has addressed critical challenges in A/B testing within ranking systems, including interference effects, budget and infrastructure capacity constraints, and bipartite experimental structures, establishing new frameworks that have fundamentally improved how experiments are designed, analyzed, and scaled. Beyond A/B testing, I develop novel methodologies that bridge observational causal inference with reinforcement learning systems, enabling more robust and efficient decision-making in production environments.

I hold a Ph.D. in Economics from Duke University. In graduate school, I was advised by Prof. Jia Li and Prof. Tim Bollerslev, working on theoretical topics in financial econometrics, with a focus on nonparametric and semi-parametric inference using high-frequency data and on testing models of financial markets.  

• Email: hyllie929@gmail.com

Curriculum Vitae (PDF)