Course Project at FoodAssemble (2026-01 – 2026-03)
- Developed a Next.js web application for recipe searching, integrating Elasticsearch and BM25 scoring logic to deliver highly accurate search results.
- Implemented a Flask backend featuring a personalized recommendation system powered by machine learning and collaborative filtering (SVD) for enhanced content discovery.
Course Project at PlayerxG (2026-01 – 2026-03)
- Developed an AI scouting engine that predicts Expected Goals (xG) by analyzing match commentary.
- Implemented a Python backend to process textual match data and generate advanced tactical metrics and deploy this project via streamlit.
Independent Project at Fraud-Scout (2026-07 – 2026-08)
- Built an AI-powered fraud detection app (Streamlit, Scikit-learn, PyThaiNLP) to identify suspicious retail transactions using NLP matching.
- Engineered a secure, in-memory data pipeline (Pandas, NumPy) to automatically detect anomalies like price manipulation and ghost transactions.
Course Project at The Social Anti-Fake News System (2025-09 – 2025-11)
- Developed an interactive, Reddit-style social discussion platform using Vue.js and Firebase, extending a basic voting system into a full forum with dynamic comment ranking based on user upvotes.
- Engineered a robust component-based backend using Java and Spring Boot to manage complex interactions (users, news threads, and votes), and successfully deployed the end-to-end application on Microsoft Azure.
