Ryan Hsiang
項達均
I am Ryan Hsiang, a senior from National Taiwan University (NTU) majoring in Electrical Engineering. My research interests lie in machine learning and its applications to physics, mathematics, and engineering. In the summer of 2025, I participated in the Summer Undergraduate Research Fellowship (SURF) at Caltech, under the guidance of Prof. Anima Anandkumar and Dr. Robert Joseph George.
Outside of research, I have extensive experience in software development through various side projects and courses. In particular, I led a group of students at NTUEE to develop the LightDance Editor, a full-stack web application for light and dance choreography built with Rust, Blender, and MySQL.
Projects
NTUEE Light Dance
Developed a full-stack web application using Rust, Blender, and MySQL to design and simulate light patterns and effects for 9 dancers and 18 props. Designed LED- and fiber-embedded costumes for each dancer, controlled via Raspberry Pi devices connected to a WebSocket server for real-time synchronization. Implemented dynamic moving light effects, such as a spinning orb and a large energy pulse, using JavaScript.
Publications
LeanDojo-v2
LeanDojo-v2 is an end-to-end framework for training, evaluating, and deploying AI-assisted theorem provers for Lean 4. It combines repository tracing, lifelong dataset management, retrieval-augmented agents, Hugging Face fine-tuning, and external inference APIs into one toolkit.
