RetailAI

A cloud retail analytics platform that predicts customer lifetime value (CLV), flags churn risk and recommends cross-sell items. Built for the Intro to Cloud Computing course at the University of Cincinnati (April 2026).

What it does

• Predicts CLV and classifies churn with Random Forest models, sorting customers into Low, Medium and High retention tiers
• Co-occurrence basket analysis over 5,000 transactions for cross-sell recommendations
• Chart.js dashboard backed by Azure SQL (3 tables, 10K+ transactions, 400 households)
• Bulk CSV upload using executemany() instead of row-by-row inserts
• Dashboard load time cut by about 80% by moving aggregation into SQL and adding in-memory caching

Architecture

CSV upload → Azure SQL → Random Forest (scikit-learn) → Flask API → Chart.js dashboard

Hosted on Azure App Service.

Tech stack

Python, Flask, Azure App Service, Azure SQL, scikit-learn, Chart.js

Run locally

git clone https://github.com/Vinay2688/RetailAI.git
cd RetailAI
pip install -r requirements.txt

Set these environment variables (your own Azure SQL database): DB_SERVER, DB_NAME, DB_USER, DB_PASSWORD, then:

python app.py

Deploy

deploy.ps1 zips the app and deploys it to Azure App Service with the Azure CLI.