A microservices-based retail analytics system that provides sales forecasting, inventory insights, and decision support using machine learning.

This project is designed to help retail businesses make data-driven decisions by analyzing sales data, predicting future demand, and generating actionable recommendations.

The system separates core API logic and ML processing into independent services for better scalability and maintainability.

• Frontend: Next.js dashboard for analytics and visualization
• Backend: Node.js + Express REST API
• AI Engine: Python (FastAPI) for forecasting and recommendations
• Database: PostgreSQL
• Cache: Redis
• Deployment: Docker + Docker Compose + Nginx

The ML service is isolated as a separate microservice to allow independent scaling and model updates without affecting backend performance.

• Sales forecasting using time-series models (ARIMA, Moving Average, Linear Regression)
• Inventory analysis with stock-level insights
• Recommendation system (restock, discount, investigate)
• Scenario simulation for revenue impact
• REST APIs for data retrieval and predictions
• Interactive dashboard for visualization

• Backend: Node.js, Express
• Frontend: Next.js, React
• AI/ML: Python, FastAPI, Pandas, NumPy, statsmodels
• Database: PostgreSQL
• Caching: Redis
• DevOps: Docker, Nginx

┌─────────────────┐    ┌─────────────────┐    ┌─────────────────┐
│   Frontend      │    │   Backend API   │    │   AI Engine    │
│   (Next.js)     │◄──►│   (Node.js)     │◄──►│   (Python)      │
│                 │    │                 │    │                 │
│ - Dashboard     │    │ - REST API      │    │ - ML Models     │
│ - Analytics     │    │ - Auth          │    │ - Forecasting   │
│ - Inventory     │    │ - Validation    │    │ - Scenarios     │
│ - Recommendations│ │ - Recommendations │ │ - Export API     │
└─────────────────┘    └─────────────────┘    └─────────────────┘
         │                       │                       │
         └───────────────────────┼───────────────────────┘
                                 │
                    ┌─────────────────┐
                    │   Data Layer    │
                    │                 │
                    │ - PostgreSQL    │
                    │ - Redis Cache   │
                    │ - File Storage  │
                    └─────────────────┘

git clone cd retail-intelligence-engine

cp .env.example .env

docker compose up --build

• Frontend: http://localhost:3000
• Backend API: http://localhost:4000
• AI Engine: http://localhost:8000

• Forecast future sales for products
• Identify low-stock or overstock situations
• Generate recommendations based on trends
• Simulate revenue impact for different scenarios

• API details: API_DOCUMENTATION.md
• Deployment guide: DEPLOYMENT.md

• Designed a microservices architecture separating ML and backend logic
• Integrated ML models into a production-style API workflow
• Used Redis caching to improve performance
• Containerized services using Docker for easier deployment

This project focuses on system design and ML integration rather than optimizing model accuracy.