Prescriptive Financial Decision Engine for Retail Credit Defense Submission for LT HackFest 2026 (FinTech Domain)
Over 70% of retail loan defaults are due to temporary cash-flow mismatches, not intentional insolvency. Yet, the financial system treats bad timing as a default.
When a borrower misses an EMI by even one day, they are hit with a ₹500+ NACH bounce fee and suffer long-term CIBIL score damage. Current systems are purely reactive (recording damage after it happens), and budgeting apps only track past spending.
Chrona AI shifts retail credit from reactive penalties to proactive optimization.
We don't just predict that a user will bounce an EMI. Our engine dynamically calculates the mathematical cost of a bounce versus the cost of an intervention (e.g., transferring a balance, pausing an SIP) and prescribes the cheapest, lowest-friction action to keep the account above zero.
The Math of a Bounce:
- ❌ Do Nothing: ₹590 Bounce Fee + 30 Point CIBIL Drop + Overdue Interest = ₹940+ Loss
- ✅ Chrona AI Action (Credit Transfer): ₹199 Processing Fee + 0 CIBIL Impact = ₹741 Saved instantly.
- 🔗 Account Aggregator (AA) Integration (Simulated): Ingests raw transaction data securely via India's consent-based DPI framework.
- 🔮 60-Day Forward Simulation: Projects the user's daily bank balance identifying exact liquidity shortfalls before they trigger a default.
- 🧠 Prescriptive "What-If" Engine: A real-time decision matrix that models the financial impact of different restructuring choices.
- ⚡ One-Click Execution: Users can instantly apply the AI-recommended strategy to restructure debt and protect their credit profile.
(Note to judges: Replace this text with a GIF or screenshot of your working UI showing the chart going from Red to Green)
Frontend (Prototype):
- React 18
- TypeScript
- Tailwind CSS (Dark Mode FinTech UI)
- Recharts (For dynamic financial time-series visualization)
- Lucide React (Iconography)
Backend / Logic Engine (Conceptual Architecture):
- Python (Pandas)
- Meta Prophet / SARIMAX (Cash-flow forecasting)
- XGBoost (Intervention optimization)
To run the frontend simulation of Chrona AI on your local machine:
- Clone the repository
git clone [https://github.com/yourusername/chrona-ai.git](https://github.com/yourusername/ChronaAi.git) cd ChronaAI