Naver-Lens: AI-Powered E-Commerce Review Summarizer

⭐️ A NAVER Vietnam AI Hackathon 2025 Project ⭐️

   

Naver-Lens is a full-stack e-commerce application that enhances the online shopping experience with an integrated AI assistant. Developed for the NAVER Vietnam AI Hackathon, it features a modern shopping interface and provides real-time, structured summaries of product reviews using NAVER's HyperCLOVA model.

💡 Note: This is an archived showcase of the project. The live NAVER Cloud deployment and HyperCLOVA AI services were exclusive to the hackathon and are no longer active. This repository has been configured with a "showcase mode" that uses a cached AI response to demonstrate full UI functionality.

🎯 My Role & Key Contributions

I was the backend architect and deployment lead on our four-person team, team-triple-threat. My key contributions included:

• System Architecture: I designed and architected the core Node.js backend from the ground up, implementing a modular MVC (Model-View-Controller) pattern to ensure a clean separation of concerns and maintainable code under a tight deadline.
• Cloud Deployment & DevOps: I provisioned and configured a Linux VM on Naver Cloud Platform from scratch. I then containerized the application with Docker and managed the full production deployment, making our project accessible.
• Strategy & Presentation: I co-developed the initial product concept, documented technical requirements, and designed the final presentation slides that we used to pitch our project.

   

🛠️ Technical Deep Dive

Backend (Node.js / Express.js)

The backend is built on a robust and scalable architecture. It handles product data management, filtering, pagination, and serves as the bridge to the AI service. Caching was implemented to optimize performance for frequently accessed product data.

🤖 AI Summarization (Server-Sent Events)

The project's core innovation is its AI-powered summary panel. When a user requests a summary, the backend streams the response using SSE. This provides a significantly better user experience than waiting for a full response, as the summary appears interactively.

   

📊 Database Schema (MongoDB)

Product data was modeled using a relational structure within MongoDB to accommodate complex product details, including multiple images, nested options, and a list of reviews. This design ensured efficient querying for both the product catalog and the AI summarization context.

☁️ Cloud Deployment (Naver Cloud Platform)

The entire application was containerized using Docker and deployed to a production environment on a Naver Cloud Platform server, demonstrating real-world deployment skills beyond local development.

✨ Visual Showcase

The UI was designed to be clean, modern, and intuitive, mirroring the aesthetic of leading e-commerce platforms. It includes light and dark modes, as well as multi-language support (English/Korean).

🇺🇸/🇰🇷 Homepage

   

🛒 Product Catalog & Details

       

🚀 How to Run (Showcase Mode)

1. Clone the repository

git clone https://github.com/wheevu/naver-lens.git
cd naver-lens

2. Run the Backend

cd backend
npm install
# Ensure SHOWCASE_MODE=true in .env
npm run dev

3. Run the Frontend

cd frontend
npm install
npm run dev

💻 Technology Stack

• Frontend: React 19, Vite, TypeScript, TailwindCSS, i18next
• Backend: Node.js, Express.js, MVC, Server-Sent Events (SSE)
• Database: MongoDB, Mongoose
• AI: NAVER HyperCLOVA (original), Mock Service (showcase)
• Deployment: Docker, Naver Cloud Platform, Linux (Ubuntu)