ABOUT ME
I am an incoming M.Sc. student in Data Science & Machine Learning (DSML) at the National University of Singapore (NUS), holding a Bachelor's degree in Physics from Nanjing University (NJU).
I became deeply fascinated by Large Language Models the moment I first used them, which served as the primary catalyst for my transition into AI. Starting in the second semester of my sophomore year, I spent a year and a half diving into computer science fundamentals, machine learning, and deep learning.
As LLM capabilities matured to power real-world applications, I pivoted my focus toward AI agents. My exploration centers on two key directions: whether AI agents can autonomously generate innovative results over long horizons toward an 'AI Scientist' — see LLM Agent Hub & Aura — and how AI and humans interact in collaborative, creative workflows — see A2Learn.
Along the way, I also mastered fundamental web development skills in Ruby on Rails and Node.js to power full-stack AI applications — see AI Course Platform.
SKILLS
PROJECTS
TALKS & BLOG
Undergraduate Thesis Defense
Nanjing University · 2026
My undergrad thesis defense at the Department of Physics, NJU.
Preview slides
NOW
- 🎓 Preparing for M.Sc. in DSML at NUS
- 🔬 Working on assignments for Deep Learning Systems (CMU 10-414/714)
- 💻 Deploying and testing AI Course Platform on the cloud
- 🏸 Desperately looking for Low-Intermediate / High-Intermediate badminton games
THOUGHTS
The flywheel is a fascinating framework for driving incredible breakthroughs. We see it in DL & RL (dataset → model → evaluation → update), in Lean Startup (build → measure → learn), and even in Popper's philosophy of science (hypothesis → prediction → explanation → new hypothesis). For the flywheel of autonomous AI, the bottleneck inevitably lies in evaluation. As we build AI agent systems, we are moving toward 'expert-in-the-flywheel'—where domain expertise becomes more vital than ever.