Recover failed payments, analyze transaction data, prioritize recovery cases, and assist recovery teams with intelligent insights.
RecoverAI provides a centralized dashboard for monitoring transactions, analyzing payment failures, managing recovery priorities, and interacting with the AI Recovery Agent.
Live Application: https://recover-ai-orcin-two.vercel.app/
RecoverAI is an intelligent revenue recovery platform designed to help businesses identify failed payments, analyze transaction and customer information, prioritize recovery cases, and support recovery teams with AI-assisted insights.
The platform combines:
- ⚛️ React frontend
- ⚡ FastAPI backend
- 🐍 Python
- 🤖 Machine Learning
- 🗄️ PostgreSQL
- 📊 Transaction analytics
- 📁 CSV data processing
- 🤖 AI-assisted recovery workflow
The goal is simple:
Turn failed payment data into actionable recovery opportunities.
A centralized dashboard to monitor payment recovery operations.
It provides visibility into:
- Total transactions
- Failed payments
- Transaction amounts
- Recovery priorities
- Customer information
- Payment analytics
RecoverAI allows users to view detailed transaction information.
- Transaction ID
- Customer ID
- Customer Name
- Amount
- Payment Method
- Failure Reason
- Attempt Number
- Previous Success Rate
- Customer Transaction Count
- Customer Lifetime Value
- Recovery Priority
Transactions can be organized according to their recovery priority.
| Priority | Description |
|---|---|
| 🔴 High | Requires immediate recovery attention |
| 🟡 Medium | Requires follow-up |
| 🟢 Low | Lower-priority recovery case |
This helps recovery teams focus their efforts on the most important transactions.
The AI Recovery Agent provides an intelligent interface for analyzing failed payment situations and supporting recovery decisions.
It can help users understand:
- Why a payment failed
- Customer transaction context
- Recovery priority
- Potential recovery actions
- Important transaction information
Users can upload transaction data through CSV files.
RecoverAI processes the uploaded data and displays the transactions inside the dashboard.
transaction_id
customer_id
customer_name
amount
payment_method
failure_reason
attempt_number
previous_success_rate
customer_transaction_count
customer_lifetime_value
priority
transaction_id,customer_id,customer_name,amount,payment_method,failure_reason,attempt_number,previous_success_rate,customer_transaction_count,customer_lifetime_value,priority
TXN001,CUST001,Rahul Sharma,12500,UPI,Insufficient Balance,2,0.82,18,95000,High
TXN002,CUST002,Ananya Singh,4500,Credit Card,Card Declined,1,0.91,12,62000,Medium
TXN003,CUST003,Aman Verma,1800,Net Banking,Timeout,3,0.65,7,28000,LowRecoverAI contains a dedicated Machine Learning module under the ml/ directory.
The ML pipeline is responsible for training models and generating predictions from transaction data.
ml/
│
├── __init__.py
├── train.py
├── predict.py
└── models/
| File / Folder | Purpose |
|---|---|
train.py |
Machine Learning model training |
predict.py |
Loading the trained model and generating predictions |
models/ |
Stores trained model files |
__init__.py |
Initializes the ML package |
- 🐍 Python
- 🤖 Scikit-learn
- 🐼 Pandas
- 🔢 NumPy
- 📊 Feature Processing
- 🎯 Predictive Analysis
Transaction Data
↓
Data Processing
↓
Feature Preparation
↓
Model Training
↓
Trained Model
↓
Prediction
↓
Recovery Analysis
RecoverAI provides analytics to help understand payment and transaction patterns.
Analytics can be used to monitor:
- Payment failures
- Transaction amounts
- Recovery priorities
- Customer activity
- High-value transactions
- Recovery opportunities
RecoverAI provides a modern responsive interface with:
- 🌙 Dark Mode
- ☀️ Light Mode
- 💻 Desktop support
- 📱 Responsive layout
- 🐍 Python
- ⚡ FastAPI
- 🔗 REST APIs
- 🗄️ PostgreSQL
- ⚛️ React
- JavaScript
- HTML
- CSS
- 🤖 Scikit-learn
- 🐼 Pandas
- 🔢 NumPy
- 📊 Feature Processing
- 🎯 Predictive Analysis
- CSV Data Processing
- Transaction Analysis
- Customer Data Analysis
- Payment Failure Analysis
- Recovery Priority Analysis
- AI-assisted Recovery Workflow
- Git
- GitHub
- VS Code
- Postman
┌──────────────────────┐
│ USER │
└──────────┬───────────┘
│
▼
┌──────────────────────┐
│ React Dashboard │
│ Frontend │
└──────────┬───────────┘
│
REST API
│
▼
┌──────────────────────┐
│ FastAPI │
│ Backend │
└──────────┬───────────┘
│
┌─────────────────────┼─────────────────────┐
│ │ │
▼ ▼ ▼
┌──────────────┐ ┌──────────────┐ ┌──────────────┐
│ Transaction │ │ Machine │ │ AI Recovery │
│ Processing │ │ Learning │ │ Agent │
└──────┬───────┘ └──────┬───────┘ └──────┬───────┘
│ │ │
└─────────────────────┼─────────────────────┘
▼
┌──────────────────────┐
│ PostgreSQL │
│ Database │
└──────────────────────┘
CSV / Transaction Data
│
▼
Transaction Validation
│
▼
Data Processing
│
▼
Customer & Payment Analysis
│
▼
ML Model / Prediction
│
▼
Recovery Prioritization
│
┌──────────┴──────────┐
▼ ▼
Priority Queue AI Recovery Agent
│ │
└──────────┬──────────┘
▼
Recovery Action
│
▼
Analytics
RecoverAI/
│
├── backend/
│ ├── main.py
│ ├── routes/
│ ├── services/
│ ├── models/
│ └── requirements.txt
│
├── frontend/
│ ├── src/
│ ├── public/
│ └── package.json
│
├── ml/
│ ├── __init__.py
│ ├── train.py
│ ├── predict.py
│ └── models/
│
├── data/
│ └── sample_transactions.csv
│
├── assets/
│ ├── recoverai-dashboard.png
│ ├── transactions.png
│ ├── ai-recovery-agent.png
│ └── analytics.png
│
├── .gitignore
├── README.md
└── requirements.txt
git clone https://github.com/Himanshu1137/RecoverAI.git
cd RecoverAIpython -m venv venv.\venv\Scripts\Activate.ps1pip install -r requirements.txtCreate a .env file in the project root:
DATABASE_URL=your_postgresql_connection_stringAdd any other required API keys or configuration values used by your application.
⚠️ Never commit.env, API keys, passwords, or database credentials to GitHub.
Start the FastAPI server:
uvicorn main:app --reloadBackend:
http://127.0.0.1:8000
http://127.0.0.1:8000/docs
Open another terminal:
cd frontend
npm install
npm run devOpen the local URL displayed by Vite.
From the project root:
python ml/train.pyAfter training, the generated model can be used for prediction through:
python ml/predict.pyAdjust the commands if your
train.pyorpredict.pyexpects specific arguments or input files.
Run backend tests:
pytestAPI testing can be performed using:
- Postman
- FastAPI Swagger UI
RecoverAI can help businesses dealing with:
- Failed online payments
- Subscription payment failures
- Recurring payment failures
- Payment retry management
- Customer payment follow-ups
- Revenue leakage
- Transaction prioritization
- Payment recovery operations
Failed payments can result in revenue loss when businesses cannot quickly identify which transactions deserve recovery attention.
Traditional dashboards often answer:
Which payments failed?
RecoverAI aims to answer:
Which failed payment should we recover first, and why?
The platform combines transaction data, customer context, Machine Learning, analytics, and AI-assisted recovery workflows to turn payment failures into actionable recovery opportunities.
| Area | Implementation |
|---|---|
| Frontend | React |
| Backend | FastAPI |
| Programming | Python |
| Database | PostgreSQL |
| API | REST API |
| Machine Learning | Scikit-learn |
| Data Processing | Pandas, NumPy |
| Input | CSV |
| AI | AI-assisted Recovery |
| UI | Responsive Dashboard |
| Theme | Dark / Light Mode |
- Advanced recovery-probability model
- Automated payment retry recommendations
- Email/SMS recovery integration
- Customer segmentation
- Recovery success prediction
- Real-time payment monitoring
- Advanced AI recovery strategies
- User authentication
- Role-based access control
- Cloud deployment
- Automated recovery reports
For production deployment:
- Keep secrets in environment variables
- Never commit
.env - Validate uploaded CSV files
- Secure database credentials
- Implement authentication and authorization
- Follow API security best practices
B.Tech Computer Science & Engineering (Artificial Intelligence)
- Artificial Intelligence
- Machine Learning
- Software Engineering
- Backend Development
- Full-Stack Development
- AI-powered Applications
If you find RecoverAI useful or interesting, consider giving the repository a ⭐ on GitHub.
This project is developed for educational, portfolio, and demonstration purposes.