Full-Stack E-Commerce Platform with AI Recommendations

Portfolio project demonstrating microservices architecture with ML-powered product recommendations.

Tech Stack

• Frontend: React + TypeScript + Redux Toolkit
• Backend: Express (Node.js) + FastAPI (Python)
• Databases: MongoDB, PostgreSQL, Redis, Milvus
• ML: Collaborative Filtering + Content-Based (HuggingFace)
• DevOps: Docker Compose

Getting Started

Prerequisites

• Docker Desktop
• Node.js 20+
• Python 3.11+

Quick Start

# Start all services
docker compose up -d

# Initialize data
docker compose exec product-service npm run seed
docker compose exec recommendation-service python -m app.ml.embed_products

# Access services
# Frontend: http://localhost:3000
# Product API: http://localhost:3001
# Recommendation API: http://localhost:8000

Development

Each service can be developed independently:

• Product Service: cd product-service && npm run dev
• Recommendation Service: cd recommendation-service && uvicorn app.main:app --reload
• Frontend: cd frontend && npm start