Skip to content

About

Resources

Stars

1 star

Watchers

0 watching

Forks

 
 

Repository files navigation

GrowthOS V2 - AI Decision Intelligence Platform for Indian Retail

GrowthOS V2 is a domain-specific, AI-assisted decision-intelligence platform built for Indian retail businesses. It transforms raw retail data into actionable business decisions using a strict diagnostic loop: Data → Diagnosis → Prediction → Simulation → Recommendation → Action → Measurement.


🌟 Key Features

  1. Executive Command Center Dashboard: Real-time Health Score (0-100), 30-day revenue trends, transaction metrics, and category revenue share distribution.
  2. Automated Root-Cause Variance Engine: Multi-dimensional variance breakdown (Volume vs. Price/Mix effect, Store drivers, Category drivers).
  3. RFM Customer Intelligence & Churn Alert: Recency, Frequency, and Monetary segmentation (VIPs, Loyal, At-Risk Churn warnings).
  4. Inventory Days of Cover (DoC) & Stockout Risk: Velocity forecasting, Days of Cover calculations, and automated purchase order reorder points.
  5. Interactive Scenario Decision Simulator: Dual-slider what-if simulator projecting 30-day revenue and gross margin impact for promotional discounts & price updates.
  6. "Ask GrowthOS" Grounded AI Assistant: OpenRouter free models API integration + deterministic grounded rule fallback engine with evidence citations.
  7. Adaptive Multi-Tenant Demo Dataset: 3 pre-seeded retail profiles (High-Tech Multi-Store, Medium Digital Kirana, Basic Traditional Store) demonstrating graceful degradation.

🏗️ Architecture & Stack

  • Frontend: React + TypeScript + Vite (Clean, crisp, light-themed professional enterprise UI; no harsh gradients).
  • Backend: Python FastAPI + Pydantic v2 + SQLAlchemy ORM (Calculations in Pandas/NumPy).
  • Database: Supabase PostgreSQL (Free Tier compatible) + SQLite fallback (growthos.db).
  • AI Engine: OpenRouter API (meta-llama/llama-3.3-70b-instruct:free) with strict evidence grounding.

🚀 Quickstart (Local Running)

1. Backend Server (FastAPI)

cd backend
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt
python3 -c "from app.seed.seed_data import seed_database; seed_database()"
uvicorn app.main:app --host 0.0.0.0 --port 8000
  • API Documentation: http://localhost:8000/docs

2. Frontend App (React + Vite)

cd frontend
npm install
npm run dev
  • Command Center App: http://localhost:5173 or http://localhost:5174

☁️ 100% Free-Tier Deployment Strategy

  • Frontend: Deploy on Vercel or Netlify (Free Static Hosting).
  • Backend: Deploy on Render (Free Web Service).
  • Database: Deploy on Supabase (Free Cloud PostgreSQL).
  • AI Models: Powered by OpenRouter Free Tier Models.

About

Resources

Stars

1 star

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages