I'm an EECS undergraduate at UC Berkeley, aspiring ML researcher, with interests on both ends of the LLM stack, from harness engineering and evaluating agents to compute-aware efficient training and inference.
Currently, I research how to make verifiable and rigorous agents for healthcare research at UCSF's Center for Intelligent Imaging.
Previously, I interned under an analyst developer at Boeing and worked on GAN research under graduate student advisment at MIT CSAIL.
UCSF Radiology, Sohn Lab
Building and evaluating agentic systems for verifiable and impact-aware biomedical research. Previously fine-tuned ViT foundation models for disease risk classification, designed automated registration for tracking subsolid nodules in lung imaging across CT scans over time
Boeing
Developed Python implementations and technical explanations of classical time-series forecasting methods for a 27,000-word introductory forecasting text under the mentorship of senior analyst developer.
GLEAN: Generative Learning for Eliminating Adversarial Noise
Trained a Generative Adversarial Network (GAN) backed with a Fast Fourier Convolutional Neural Network (repurpoused from image-to-image super-resolution upscaling tasks) for removing adversarial perturbations from artwork images
A generative framework for studying the limits of adversarial protections for digital artwork.
A practical introduction to statistics, Python, and machine-learning approaches to forecasting.
Rose Hills Summer Research ScholarshipAwarded $5,000 for machine learning research in GANs for the prediction of lung disease progression
B.S. Electrical Engineering & Computer ScienceUniversity of California, Berkeley