// Cornell University — Statistics · CS · Product Strategy · Dec 2026
I like to build fast and iterate faster. SwiftUI apps, neural networks, autonomous robots — if it ships, I'm in. Currently exploring how to orchestrate agentic AI workflows that balance autonomous reasoning with token efficiency.
01 — About
I'm a Statistics + CS student at Cornell University, graduating Dec 2026. My work spans iOS development, machine learning research, and teaching — built on a foundation of discipline from serving in the U.S. Marine Corps Reserves.
At Cornell App Development, I build iOS apps used by the Cornell community using SwiftUI and TCA. I also instruct CS 1998, Cornell's student-run iOS course, where I redesigned the curriculum and manage a team of 11 TAs.
I've done ML research at Cornell's Architecture Robotics Lab and as a STARS Fellow at Cal Poly Pomona, working on transformer attention mechanisms and autonomous robot control.
Interests: Archery, Entrepreneurship, Robotics, Sci-fi, Psychology.
Important Times
Started somewhere far from here.
Nine years old. New country, new language, figured it out.
Early mornings, minimum wage. Learned what it means to work.
Went to 3 different ones. It wasn't the right path. Took time to find one that was.
Tables, tips, and a lot of time to think about what comes next.
COVID hit. Moved anyway. Sometimes a reset is the move.
Started from scratch academically. Ended with a 4.0.
MCRD, SD. Thirteen weeks. A different kind of education.
Accepted my dream school UCLA and UC Berkeley as a Regents scholar. Chose Cornell! Dropout to Ivy. The long way around is still a way.
Agentic AI, Cornell App Dev, instructing students.
02 — Education
03 — Experience
04 — Selected Projects
Book trading app built with SwiftUI featuring modular architecture and explicit state management. Collaborated cross-functionally to ship a production-ready app under hackathon time constraints.
Designed a Siamese Neural Network with ResNet backbone achieving 88% accuracy — outperforming logistic regression and random forest baselines. Implemented full data preprocessing, cross-validation, and error analysis pipeline.
Analyzed Depop transaction data as project lead to study pricing dynamics and seller behavior. Engineered features for brand popularity and time-to-sale; applied regression models to identify drivers of sell-through and price dispersion.
Productivity platform integrating Pomodoro timers and task tracking. Designed session-level analytics to measure user behavior patterns and surface engagement bottlenecks.
05 — Toolkit
Now
// Open to opportunities — internships & full-time