University of Minnesota Twin Cities
Computer Science · Mathematics · Quantitative Economics
- GNN101 - an interactive visualization tool for graph neural networks.
- Agentopia - a framework that helps you understand LLM behaviors in multi-agent systems.
- SciDER - an AI scientist that integrates data-centric principles and self-evolved memory.
- gnn-exp - a Python package for graph neural network visualization.
- LargeGraphs.jl - a webGL-based package for large graph rendering in Julia.
- DigitalDrifting - a web visualization for digital growth identification in the US.
- ScholaVIS - a visual analytic toolkit for scholarly writing behaviorial analysis.
- Kandor - an LLM-based simulation engine for world building with a temporal KG memory bank.
Hi! I'm Yilin (Harry) Lu, an undergraduate researcher at the University of Minnesota Twin Cities, pursuing a triple major in Computer Science, Mathematics, and Quantitative Economics, with a minor in Statistical Science (Astro-Informatics).
My research interests mainly focus in two directions: (1) develop interactive systems to help non-AI experts make sense with AI; (2) develop Human-AI collaborative technologies for scientific discovery.
My Research Experience:
-
Research Intern (Fully-funded) — University of Pennsylvania (incoming)
- Advisor: Prof. Andrew Head
- Topic: Interactive Systems for Scientific Reasoning and Research
-
Research & Engineering Lead — University of Minnesota (Jan 2024 - Present)
- Advisor: Prof. Qianwen Wang
- Topic: AI Sensemaking for Non-expert Users
-
Research Intern — William & Mary (Nov 2025 - Present)
- Advisor: Prof. Qingyun Wang
- Topic: Agentic Memory for Scientific Discovery
-
Research Intern — Minnesota Institute for Astrophysics (Sept 2023 - May 2024)
- Advisors: Prof. Patrick Kelly & Dr. Mandeep Gill
- Topic: Crowdsourcing Systems for Scientific Discovery
Before college, I studied Mathematics at San Diego State University through the Dual Enrollment Program and graduated from Torrey Pines High School (CA).
Human-Computer Interaction, Explainable AI, Visualization, AI Agents, AI for Scientific Discovery.
Technologies I frequently use: D3.js, LangGraph / LangChain, Jupyter Widget Development, Python Data Science Stack.