I study computer science and statistics at Harvard University. I am passionate about developing AI systems, building applications, and creating value from data. I love music and language, too.
I am an undergraduate AI researcher at the Kempner Institute, where I am working on multi-agent reasoning with language models under Prof. Yilun Du and Prof. Sham Kakade. I am also collaborating with Research Fellow Aaron Walsman to build mechagogue, a pedagogical reinforcement learning repository with implementations of classical algorithms in JAX.
Currently, I am a Software Engineer Intern in Institutional Securities Technology at Morgan Stanley. During the year, I direct full-stack development at The Harvard Crimson. You can also find me with the Bach Society Orchestra.
Previously, I was a Machine Learning Engineer Intern at FADEL and an AI Research Intern at the Slade Lab.
Some of my experiences while working on mechagogue, a reinforcement learning repository with from-scratch JAX implementations of classic RL algorithms.
The official web app for Harvard's student-run movie theater, with a proprietary content management system, seat booking, archives, merch shop, and more.
Prune As You Tune (PAYT) interleaves pruning of pre-trained parameters with LoRA fine-tuning updates guided by a custom knowledge distillation loss function. PAYT can achieve up to 50% sparsity with minimal accuracy degradation and lower perplexity on the original task compared to baselines such as full fine-tuning and prune-then-fine-tune methods.
An investigation of the effect of neural network pruning on the explainability of LLMs, with the goal of understanding how model compression techniques may affect AI interpretability.
Training and evaluation of diffusion models on the AddBiomechanics dataset for generating sequences of human motion. Find our lightning talk at the 2025 Harvard Generative AI Symposium here.
