Software Engineer · University of Toronto CS
Currently at DoorDash, engineering the core event-driven systems that drive intelligent notification dispatching, targeting, and deliverability.
Parameterized checkpoint restoration configurations in the core streaming runtime to significantly improve state recovery flexibility.
Resolved a critical query validation bug where trailing zeros in temporal literals broke SQL filter pushdowns to BigQuery.
Contributed core visualization modules to a highly adopted Python memory visualization package with over 39,000 downloads.
Contributed feature enhancements and performance optimizations to a full-stack learning management application serving 10,000+ students.
A backend-as-a-service platform for developing web applications with AI agents, featuring instant zero-config deployment, a managed database, and real-time websockets.
A productivity tool that parses course syllabi using Google Gemini and generates dynamic schedules.
A real-time chat platform that translates messages across 11 languages using the DeepL API, with a Java backend and MongoDB.
An automation pipeline that generates TikTok videos using OpenAI Whisper for transcription, Reddit for content, and FFmpeg for editing.
A social media analytics app that identifies unfollower trends from Instagram JSON exports.
A full-stack web app that displays random cities enriched with Wikipedia data and allows users to submit star ratings.
A recreation of Nintendo's Dr. Mario game written entirely in MIPS assembly, featuring gravity mechanics and sprite-based animation.
A developer dashboard with a Flask backend and Next.js frontend that communicates with an STM32 via UART for real-time monitoring.