Hello!

I’m an applied machine learning researcher and systems modeler passionate about using generative AI and causal modeling to understand and improve complex socio-technical systems. My expertise spans natural language processing (NLP), simulation modeling, and decision-support systems, with a focus on leveraging large language models (LLMs) for knowledge extraction, causal reasoning, and automation.

Across my work, I’ve developed scalable AI pipelines that translate messy, real-world data—including text and behavioral logs—into interpretable, actionable insights. My research continues to explore how human-aligned AI can be used to design safer, fairer, and more intelligent decision-support systems.

I hold a PhD in Industrial and Systems Engineering and a Master’s in Computer Science, both from Virginia Tech. During my PhD at the Systems Performance Lab, I applied supervised and unsupervised machine learning (Random Forest, XGBoost, ANN), optimization techniques (DEA), and system dynamics modeling to analyze cognitive workload and extract patterns from daily operational data that captured workload and decision behaviors of operators in traffic control centers.

As part of my PhD, I also spent a year as an AI Resident at GoogleX, where I explored how LLMs can automate causal loop diagram (CLD) generation to scale systems thinking and support complex model-building tasks.

Currently, I’m a Postdoctoral Research Fellow at Harvard Medical School, based at Massachusetts General Hospital’s Institute for Technology Assessment and MJ lab, where I apply LLMs and NLP to health interview data to improve decision modeling and evaluate healthcare system performance.

📫 Feel free to contact me at gliu26@mgh.harvard.edu