
A comprehensive art education platform enabling online course registration, event management, and ticketing with real-time features. Features include a full admin CMS for content management, voting and polling system for student engagement, advanced SEO optimization, multi-language support, and comprehensive analytics dashboard.

A modern corporate website showcasing construction projects and news with a dynamic CMS, optimized for performance and SEO. Includes comprehensive admin CMS for content management, interactive project gallery with image optimization, multi-language support (Vietnamese/English), responsive design, and advanced SEO features.

An advanced AI-powered multimodal search system for fashion products using state-of-the-art vision-language models (CLIP, BEiT-3). The system supports natural language queries to find visually similar fashion items, implements scalable distributed architecture with Kafka for real-time data streaming, Apache Spark for large-scale data processing, MongoDB for flexible document storage, and FAISS for efficient similarity search. Features include image-to-image search, text-to-image retrieval, and hybrid search capabilities.

A modern, responsive portfolio website with a full-featured blog system, project showcases, and an admin dashboard. Built with Next.js, featuring 3D animations, internationalization, rich text editor for content management, and comprehensive blog functionality with tags, categories, and search.
CTalk Panel Dashboard is a web application for managing and visualizing survey panel data. The application allows viewing panel members overview, managing projects, and analyzing respondent data. The dashboard supports multiple languages (Vietnamese/English), processes data from Excel/Google Sheets, displays statistical charts (bar charts, donut charts), and integrates PDF report viewing. The application provides filtering, search, and sorting features for project data with a responsive and modern interface.

A Python-based recommendation system integrating Collaborative Filtering and Content-Based Filtering algorithms. Built with Tkinter GUI, using MovieLens 100k dataset, the system provides movie recommendations through user-user and item-item collaborative filtering, as well as content-based recommendations using TF-IDF vectorization and cosine similarity. Features include real-time recommendations, sparse matrix operations for efficient memory usage, and a modular architecture.

A modern full-stack blog system with Next.js frontend and .NET 9.0 backend. Features include comprehensive CMS with post and category management, Clerk authentication, Cloudinary image storage, rich text editor (TipTap), multi-language support (i18n), dark mode, SEO optimization, and admin dashboard. Includes interactive features such as comment system, voting/like functionality, and real-time notifications. Built with clean architecture, Entity Framework Core, SQL Server, JWT authentication, and Docker containerization.

A modern e-commerce platform specializing in pure honey and honey-based products. Built with Next.js 14, featuring product catalog, shopping cart, order management, and admin dashboard. Includes Clerk authentication, Prisma ORM with PostgreSQL (Supabase), Cloudinary image storage, Chart.js analytics, Sentry error tracking, Resend email service, and multiple payment methods (COD, Bank Transfer, MoMo). Features responsive design, page view tracking, and comprehensive product management system.