A website that can simulate history weather case. Providing simulation time, tools, questions and answer block for student to complete the exercise though the browser. Also include an admin panel for admin to manage the weather case, no need to handle the creation / modify by writing code.

Creating a personal website to showcase my projects, skills and experiences. This website is built with Next.js, TailwindCSS and Typescript in SSG mode.

A request-for-quote (RfQ) system for Gold Label Tech Ltd. Using Next.js with Prisma and MariaDB. Enhance on SEO with SSR and also include Admin page for managing the RfQs.

A RFID gate system with human direction detection capabilities. The system is built with Spring Boot as the main server, connected to an RFID reader via plain sockets, and utilizes FastAPI with WebSocket for YOLO-based detection. It can detect human movement directions using YOLO and record the corresponding item access directions.

An intelligent chatbot leveraging Retrieval-Augmented Generation (RAG) to provide accurate and context-aware responses. Built with a React frontend and FastAPI backend, it integrates seamlessly with AWS Bedrock to enhance conversational capabilities.

A comprehensive Point of Sale (POS) and Inventory Management system supporting dark/light modes and English/Chinese locales. Built with Nuxt3 on the frontend using TailwindCSS, Element-Plus, and Tabulator-table, and Python Django REST framework with Maria DB and Django ORM on the backend. Ensures database transactions adhere to ACID properties.

A comprehensive inventory management system supporting dark/light modes and English/Chinese locales. Built with Vite and Vue3 on the frontend using Ant Design and Tabulator-table, and Python Django REST framework with MySQL and Django ORM on the backend. Ensures database transactions adhere to ACID properties.

This project proposes an integrated AI-powered solution featuring a web-based user portal for speakerspecific dubbing and translation. The system processes input videos in a source language (e.g., English) and generates translated output (e.g., Cantonese) while preserving the original speakers' vocal characteristics, including voice and tone.

The AI Calories Helper App is a cross-platform mobile application designed to assist users in tracking and managing their daily caloric intake with the help of artificial intelligence. Developed using Expo (React Native) with TypeScript for the front end and FastAPI for the back end, the app leverages a GPT-4o-Mini model to analyze meal images and provide nutritional information.

The project is to enhance functionality of the AI agent (OS-Copilot). I have enhanced the base prompt and added some new features, such as support Windows environments with Batch file excuation, extra momory slot for previous sub-task, previous chat history memory, and some new tools.

This study combines Twitter sentiment analysis and clustering with deep learning to identify patterns and predict market movements. By grouping stocks based on volatility and momentum, and integrating real-time sentiment signals, we create machine learning-driven portfolios optimized for dynamic trading environments.

Our project focuses on Remote Photoplethysmography (rPPG), an economical method to extract heart rate from facial videos. Key objectives include heart rate extraction, IoT-based video data retrieval, wireless IoT-server connectivity, MQTT for data communication, data visualization, and system evaluation.

A POS for restaurant with backend management system. Include a web-based application for online ordering and payment. Backend system is built by Java Swing with Orcale SQL. Web-Base Application is built by Python Flask.