Skip to content

gowthamrdyy/CredAxis

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
 
 
 
 
 
 
 
 
 
 

Repository files navigation

CredAxis - Agentic AI Personal Loan Prototype

CredAxis is a prototype demonstrating an end-to-end personal loan journey using Agentic AI. It features a conversational frontend and a multi-agent backend orchestration system.

Features

  • Conversational UI: Chat-based loan application.
  • Multi-Agent Backend: Sales, Verification, and Underwriting agents.
  • Mock APIs: Simulated CRM and Credit Bureau data.
  • Real-time Decisioning: Deterministic rules for loan approval.
  • Sanction Letter Generation: PDF generation for approved loans.

Tech Stack

  • Frontend: React (Vite), Tailwind CSS
  • Backend: Node.js, Express
  • PDF Generation: pdf-lib

Prerequisites

  • Node.js (v14+)
  • npm

Setup & Run

1. Backend

cd server
npm install
node index.js

Server runs on http://localhost:3000.

2. Frontend

Open a new terminal:

cd frontend
npm install
npm run dev

Frontend runs on http://localhost:5173 (or similar).

Demo Script

  1. Open the frontend URL.
  2. Click "Start Loan Chat" (or just start typing).
  3. Sales Agent:
    • Name: Enter any name (e.g., "Anil Kumar").
    • Phone: Use +919900112233 (Mock CRM data).
    • PAN: Use AAAAA1111A (Mock Credit data, Score 740).
    • Income: 50000.
    • Amount: 300000.
    • Purpose: "Personal use".
    • Tenure: 24.
  4. Verification Agent: Automatically verifies using mock CRM.
  5. Underwriting Agent: Checks credit score and approves/rejects.
    • If approved, a "Generate Sanction Letter" button appears.
  6. Sanction Letter: Click to generate and download the PDF.

Mock Data

  • Phone: +919900112233 (Verified Customer)
  • PAN: AAAAA1111A (Credit Score 740 - Good)
  • PAN: BBBBB2222B (Credit Score 660 - Average)

License

MIT

About

Instant, Autonomous Personal Loan Processing Powered by Multi-Agent AI

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published