My name is Pingfan Hu, a PhD candidate in Systems Engineering at George Washington University, advised by Professor John Helveston. My work focuses on data science, quantitative modeling, and software development for human behavior and transportation systems. I develop open-source software for survey design and discrete choice modeling, and build full-stack research applications with R, Python, Quarto, Shiny, and PostgreSQL. I also apply agentic AI engineering to design and deploy LLM-powered workflows that automate complex research and development tasks.
- Data Science: I study human behavior in transportation systems using discrete choice models and scenario simulation, focusing on smart charging adoption and economy to generate policy suggestions for policymakers and utilities.
- Open-Source Software: I develop open-source R tools for survey research, including surveydown, a programmable, markdown-based survey platform built on Quarto, Shiny, and PostgreSQL, and its companion GUI sdstudio.
- Agentic AI Engineering: I build reproducible workflows around Claude Code with plugins, skills, MCPs, and hooks. I document the tutorials and my own workflows in my blog, and run a workshop on GW TAI for researchers and developers.