A full-stack web platform to help Princeton students connect with each other for friendships or relationships. TigerMatch integrates a questionnaire-based matching algorithm, user profiles, photo uploads, anonymity settings, crush features, E2E encrypted messaging, and mock CAS authentication. Developed using React, TypeScript, Node.js, Express, and PostgreSQL, this platform aims to reduce isolation on campus and foster meaningful connections.
Projects
Collection of AI-powered scripts that achieve superhuman performance in GamePigeon games. Features include a world record-breaking (321,900 points) WordHunt solver using OCR and PPO algorithms, a precision Darts AI that can hit any target perfectly, and an Anagrams solver. Built with advanced computer vision and reinforcement learning techniques for automated gameplay optimization – completely hands free. Requires iPhone mirroring through macOS 15 Sequoia.
AI-powered campus (Today - Tay) assistant serving 450+ daily users, featuring real-time geospatial integration and optimized email processing. Built with Python and integrated with Google Places API. The system provides instant responses to campus-related queries, helping students navigate academic life more efficiently.
Real-time chat application with secure rooms and instant messaging, now natively integrated into this website. Features WebSocket communication, user authentication, and persistent message history. The application supports multiple chat rooms, real-time message delivery, and secure user sessions. Try it through the chat icon in the navigation bar!
Collection of advanced data structures and algorithms implementations including percolation analysis, KD-trees, WordNet, and seam carving. Built as part of Princeton's COS 226 course. The implementations focus on efficiency and practical applications in real-world scenarios.
Web application to analyze Instagram followers and following data,
identifying non-mutual connections using a C-based analyzer hosted on
Heroku. The system processes large datasets efficiently and provides
detailed analytics about follower relationships.
I initially wrote this during my COS 217 (Intro. to Programming Systems)
class to force myself to get more familiar with C instead of studying or
reading that boring asf textbook oop... the hardest part was deploying to
Heroku on C despite them having a separate architecture.
A smart door control system designed for dorm rooms; utilizing an Arduino Nano ESP32 Connect, L298N motor driver, and a DC motor to provide web and Bluetooth-controlled (BLE) door automation. The system integrates with Arduino IoT Cloud for control and status monitoring through a web interface, featuring real-time status updates and secure access control.