Personal research journal
Wesley Lu
Welcome to My Page
Hi, I research/work on a range of problems, from performance and system optimization to language models and machine learning. Some future work I am interested in is working on high-performance infrastructure that turns data into useful signals or ML/LLM systems that actually hold up in production. In the long term, I am especially interested in focusing on research around LM or ML.
I’m a CS + Math student at Georgia Tech, focused on Systems Architecture and Intelligence in Computer Science, and Probability & Statistics in Mathematics. Previously, I spent the summer at Amazon.com working on the backend for a conversational AI that helps users interact with their wishlists. I evaluated research ideas such as CoT reasoning and agent-based interaction, and I also experiment with different LangGraph architectures to improve reliability and performance. Lately, I have been exploring LM research, advised under Dr. Li and Dr. Du.
Before that, I immersed myself in fundamental equity research, personal portfolio management, and reselling ventures, generating over six figures.
currently reading: world models and some other stuff
Viewed by 987 humans
Recent
Every Interface Is Now an Agent Interface: Computer Use Agents
Training Verifiers Through Debate: Three-Agent Prover-Verifier Game
A three-agent extension of the prover-verifier game where two adversarial provers debate to train a more robust verifier.
ML Undergraduate Researcher · University of Virginia
Under the guidance of Hongru Du, PhD, I’m building an LM for human decision modeling pipeline that generates interpretable decision rules from primitives and iteratively improves them using model feedback. I also incorporate prior probabilities and update them with observed outcomes on a mobility dataset, using gradient descent to better fit parameters and calibrate predicted choice probabilities.
World Models & Vision-Language-Action Models: An Advanced Field Guide
A technical survey of how machines learn to simulate and act in the physical world from Sutton's Dyna to pi-0 and Dreamer 4 covering architectures, math, and open problems.
Jump's Crypto Reading List (ip)
This is a collection of notes from Jump's Crypto Reading List
Software Development Engineer Intern · Amazon
Incoming 2026 summer