I'm a full-stack & data engineer currently pursuing my MS in Data Analytics Engineering at Northeastern University and a UC Davis CS alum. My work bridges software engineering, machine learning, and sports analytics — with a strong interest in building tools that are both technically sound and deeply human-centered.
- ⚽ 343 Football – Building a football analytics startup aimed at delivering machine learning–driven insights on players and teams through blog posts and visual breakdowns. Each post will link to the underlying GitHub repo. Future plans include an AI assistant trained on player data to answer football questions in real time.
- 📚 Miras Uyghur Digital Archive – Building a digital archive for Uyghur cultural preservation with OCR pipelines, searchable metadata, and a full-stack Supabase backend.
- 🧠 Penalty Kick ML Modeling – Developing machine learning models to predict optimal penalty shot placement, likely shooter behavior, and goalkeeper dive direction using GAIT analysis and historical penalty to model conversion trends and decision patterns.
- 🛰️ NUHorizon: Embedded Software Engineering – Implementing advanced onboard image compression and data handling systems for a CubeSat project.
Languages: TypeScript, Python, SQL, Swift, C++, R
Frameworks: React, Next.js, Node.js, Prisma ORM, Scikit-learn, Tailwind
Tools: Docker, Supabase, Firebase, Tableau, AWS, PostgreSQL, Git, Jupyter
| Project | Description | Tech |
|---|---|---|
| Injury Prediction System | End-to-end ML pipeline for injury risk using SHAP, TransferMarkt scraping, and feature engineering | Python, R, SQL, PostgreSQL |
| E-commerce Platform | Dual-app architecture with Stripe checkout & admin dashboard | Next.js, Prisma, MySQL |
| Reddit Uyghur Sentiment Analysis | NLP-driven topic modeling and sentiment tracking across 16,000+ Reddit posts. Includes DistilBERT and t-SNE visualizations. | Python, SpaCy, Transformers |
| Amazon Review Sentiment Classifier | Traditional ML pipeline for classifying 34K+ reviews using TF-IDF, SMOTE, and ensemble modeling. | scikit-learn, NLTK, Python |
| Miras Archive (Private, In-Development) | Full-stack platform for Uyghur heritage records w/ searchable metadata & digitization | Next.js, Supabase |
- 📍 Based in Boston (California-raised, Japan-born, Uyghur roots)
- 📚 Always learning — blending computer science, culture, and advocacy
- ⚽ Off the keyboard: soccer addict, gym rat, and manga enthusiast (One Piece fan)
Thanks for stopping by! 🚀
If you're hiring or collaborating on something exciting — I'd love to connect.