An open API service indexing awesome lists of open source software.

https://github.com/jitendragangwar123/seiagentpay

Empowering Merchants and Consumers with AI-Driven Crypto Payments for a Frictionless Financial Future
https://github.com/jitendragangwar123/seiagentpay

ai-agent defi etherjs foundry nextjs payment-gateway sei-network solidity walletconnect

Last synced: 17 days ago
JSON representation

Empowering Merchants and Consumers with AI-Driven Crypto Payments for a Frictionless Financial Future

Awesome Lists containing this project

README

        

# 🚀 SeiAgentPay

## 📌 Problem Statement
Traditional payment systems lack transparency, efficiency, and security, especially when it comes to stablecoin payments. Businesses and users face challenges such as:
⚡ High transaction fees
🔗 Lack of interoperability
🐢 Slow processing times

## 💡 Solution
**SeiAgentPay** is a **decentralized stablecoin payment system** that allows users to make payments seamlessly using **DAI, USDT, and USDC**. It integrates an **AI-powered recommendation system** to suggest the best stablecoin based on transaction history and user behavior, ensuring **cost-effective and efficient** transactions.

## ✨ Features
✅ **🔍 Stablecoin Recommendation** – AI-powered system to suggest the best stablecoin for payments.
✅ **💰 Multi-Stablecoin Support** – Supports **DAI, USDT, and USDC** transactions.
✅ **🔐 Secure Transactions** – Trustless and transparent payments using **smart contracts**.
✅ **🏪 Merchant Payments** – Enables **direct stablecoin transfers** to merchants.
✅ **⚡ Fast & Low-Cost** – Built on **Sei Network** for **high-speed** and **low-fee** transactions.
✅ **🔧 Admin Control** – Allows updating stablecoin addresses and managing settings securely.

## 🏗 Stablecoin Recommendation Model
SeiAgentPay integrates a **machine learning-based stablecoin recommendation model** that suggests the best stablecoin for a given transaction based on:
- **User Transaction History** (average volume & frequency)
- **Merchant Category**
- **Transaction Purpose**

### **Model Training & Deployment**
- The **RandomForestClassifier** is used to train a model with transaction data.
- The model is exposed via a **FastAPI backend** that provides an API for recommendations.
- Encoders ensure categorical data (merchant category & transaction purpose) is processed correctly.
- The recommendation engine runs on **Google Colab** for model training and can be deployed locally or on a server.

## 🚀 Getting Started

### **1️⃣ Clone the Repository**
```sh
git clone https://github.com/jitendragangwar123/SeiAgentPay
cd SeiAgentPay
```

### **2️⃣ Install Backend Dependencies**
```sh
cd stablecoin-recommender
pip install -r requirements.txt
```

### **3️⃣ Train the Stablecoin Recommendation Model (Optional)**
If you want to retrain the model, run:
```sh
python train_model.py
```
This generates updated **.pkl** model files and user transaction history.

### **4️⃣ Run the Backend API**
```sh
python main.py
```

### **5️⃣ Install Frontend Dependencies**
```sh
cd front-end
npm install
```

### **6️⃣ Run the Frontend**
```sh
npm run dev
```

## 📡 API Endpoints
The backend provides the following endpoints:
| Endpoint | Method | Description |
|-----------|--------|--------------------------------------------|
| `/` | GET | Check API status |
| `/users` | GET | Fetch user transaction data |
| `/recommend` | POST | Get the best stablecoin recommendation |

## 🤝 Contributing
We welcome contributions from the community! 🌍
Feel free to **open issues**, **submit pull requests**, or **suggest new features**.

## 📜 License
This project is licensed under the **MIT License**. 📝

## Contact Us
Please reach out to us at [email protected] or [email protected]