I'm a Software Engineer with a Master's in Computer Science (AI Specialization) from Southern Methodist University. I focus on Full Stack Development and Web Development, and I like applying ML/AI where it improves the product.
- Modern UI/UX: dark mode toggle, scroll animations, hover interactions, scroll progress indicator, and mobile navigation
- Project cards: each project includes Description / Problem / Solution / Tech Stack
- Resume section: embedded PDF preview + download link on the site
- Languages: Python, JavaScript, TypeScript, HTML/CSS, Java, Swift, MATLAB, C, Objective-C, C++, SQL
- Frameworks/Libraries: React, Next.js, React Native, Node.js, Spring Boot, Flask, FastAPI, Vite
- Data Science/ML: pandas, NumPy, XGBoost, scikit-learn, Kaggle API
- Databases: PostgreSQL, MySQL, SQLite, MongoDB, Supabase
- Tools/Platforms: Git/GitHub, VS Code, IntelliJ IDEA, Google Calendar API
- Cloud/Deployment: AWS, Vercel, Railway, Render
-
Planno — Scheduling Platform | Code | Live Demo
- Description: Scheduling platform that creates shareable booking links for interviews and meetings.
- Problem: Manual coordination leads to conflicts, missed appointments, and back-and-forth communication.
- Solution: Real-time availability via booking links + Google Calendar sync + Apple Calendar downloads.
- Tech Stack: Next.js, TypeScript, React, Supabase, PostgreSQL, Google Calendar API, Vercel
-
TastyFood — Food Ordering Platform | Code | Live Demo
- Description: Direct online ordering platform for local restaurants.
- Problem: Restaurants lose 15–30% revenue to third-party services and lose customer/pricing control.
- Solution: Role-based ordering dashboards to reduce fees and keep customer relationships in-house.
- Tech Stack: React, Vite, Java, Spring Boot, PostgreSQL, SQLite (prototype), Railway
-
Premier League MVP Predictor — ML + Analytics Dashboard | Code | Dashboard Page
- Description: ML system that predicts top MVP candidates using comprehensive player performance metrics.
- Problem: MVP selection is often subjective and can miss all-around contributors.
- Solution: XGBoost regression predicts player value from stats (goals, assists, clean sheets, influence, creativity, threat, etc.) and surfaces the top candidates via an embedded analytics dashboard.
- Tech Stack: Python, FastAPI, XGBoost, scikit-learn, pandas, NumPy, Kaggle API, Render, Looker Studio (embedded dashboard)
- Portfolio: dimitrilavin.com
- LinkedIn: linkedin.com/in/dimitrilavín
- Email: dimitrilavin@gmail.com