CredTech – ML-Based Credit Scoring Backend

Overview

CredTech is a machine-learning–driven credit scoring system that predicts credit scores based on financial and behavioral parameters.
The system leverages XGBoost for accurate predictions and exposes the model through a FastAPI backend, fully containerized using Docker for consistent deployment, maintainability, and cross-environment reproducibility.

This repository contains the backend service responsible for real-time credit score computation via RESTful APIs.

---

Here is the deployed version : https://credtech-frontend-d2jc.vercel.app/ (Note : The hosted server will be shut down after a while due to resources constraint)

Features

• XGBoost-based credit scoring model
• FastAPI-powered REST APIs
• Real-time score generation
• Dockerized for easy deployment
• Environment-independent execution
• Clean and modular backend structure

---

Tech Stack

• Python
• FastAPI
• XGBoost
• Uvicorn
• Docker
• Pydantic

---

Project Structure

credtech_backend/
│
├── main.py            # FastAPI application entry point
├── test.py            # Model testing / experimentation
├── requirements.txt   # Python dependencies
├── Dockerfile         # Docker configuration
├── .env               # Environment variables
└── README.md

---

Running the Project Locally (Without Docker)

git clone https://github.com/ritgit24/credtech_backend.git
cd credtech_backend
python -m venv venv
venv\Scripts\activate(Windows) / source venv/bin/activate(Linux)
pip install -r requirements.txt
uvicorn main:app --reload

Running with Docker (Recommended)

docker build -t credtech-backend .
docker run -p 8000:8000 credtech-backend

Access the application

API Base URL: http://127.0.0.1:8000

Swagger UI: http://127.0.0.1:8000/docs