Digital Move — Retention Intelligence Platform

ThunderHacks 2026 · Gold Sponsor Challenge

Stack

• Backend: Python · Flask · scikit-learn · pandas · numpy
• Model: LogisticRegression with StandardScaler pipeline · 5-fold cross-validation
• Frontend: Vanilla JS · Chart.js · EmailJS
• AI Emails: GPT-4o via OpenAI API
• Email Delivery: EmailJS (pre-configured)

Setup (2 minutes)

1. Install dependencies

cd digital_move
pip install -r requirements.txt

2. Add your OpenAI API key Open templates/index.html and replace:

const OPENAI_KEY = "YOUR_OPENAI_API_KEY_HERE";

with your actual key.

3. Run the app

python app.py

4. Open in browser

http://localhost:5000

Features

• ✅ Real scikit-learn LogisticRegression model (not hardcoded scores)
• ✅ 5-fold cross-validated accuracy displayed on Model page
• ✅ Dynamic customer journey mapping (auto-placed by model)
• ✅ CSV import → model retrains on new data automatically
• ✅ GPT-4o generates personalized retention emails
• ✅ EmailJS sends emails live during demo
• ✅ Live second-by-second ticker
• ✅ Export report as CSV

CSV Import Format

name,segment,mrr,last_contact,renewal,plan
Acme Corp,enterprise,4500,30,45,Enterprise 10G

Columns: name (required), segment, mrr, last_contact, renewal, plan

Built With

• emailjs
• flask
• javascript
• numpy
• pandas
• python
• scikit-learn