I am majoring in Computer Engineering at Boston University, with a concentration in Machine Learning and a minor in Business Administration & Management. I am passionate about building innovative projects that bridge the gap between technical complexity and real-world impact.
My experience spans software development, hardware engineering, and advanced research. Currently, I am working at Livelo, where I am building software to improve the customer experience.
- Currently working at Livelo, a global bike rental company based in Sydney, Australia
- Formerly worked as an Undergraduate researcher at BU's PsychoLinguistics Lab and H2X Lab
- Based in Boston, MA · Connect with me on LinkedIn
- Building software to improve the customer experience @ Livelo
An AI-powered news platform that collects, deduplicates, clusters, and summarizes breaking news from across the web to mitigate media bias and show the complete picture of an event. Built on a 3-person team.
Key Features:
- Web-scraping pipeline collecting and normalizing 500+ articles daily
- Vector-similarity clustering algorithm to group daily articles by topic
- AI-powered clustering and headline generation with the OpenAI API
- Interactive D3.js node visualization for exploring news clusters
Tech Stack: Python TypeScript React Supabase D3.js OpenAI API
GitHub: View Project | Devpost: View Submission
A deepfake verification system with a Rust-based reverse proxy that decouples media delivery from deepfake scoring.
Key Features:
- Diagnosed a self-hosted deepfake detector's 0% recall failure and replaced it with a CLIP-based classifier
- Tuned the decision threshold to reach 94% recall and 79% accuracy on a 100-image test set
- Rust reverse proxy resolving a verification badge in ~2,000ms via an asynchronous background job
- Results synced through Supabase
Tech Stack: Rust TypeScript React WebAssembly Supabase Cloudflare Workers
