AI Agent E-commerce Backend

An AI agent powered e-commerce backend built with FastAPI.
This project demonstrates how an AI agent can interact with backend APIs using tool calling to perform e-commerce related tasks.

The system uses a LLM (Ollama) to decide which tools to call.

---

Features

• AI Agent with tool calling
• LLM integration using Ollama (local or cloud models)
• FastAPI backend service
• Order management API
• Product query API
• Example AI-agent interaction workflow

---

Project Structure

ai-agent-ecommerce-backend/
│
├── backend/
│ └── server.py
│
├── agent/
│ └── agent.py
│
├── tools/
│ └── order_tools.py
│
├── models/
│ ├── order.py
│ └── product.py
│
├── tests/
│ └── test_api.py
│
├── requirements.txt
└── README.md

---

Architecture

User
 ↓
FastAPI API (/agent)
 ↓
LLM (Ollama)
 ↓
AI Agent Logic
 ↓
Tool Functions
 ├ get_products()
 ├ get_order()
 └ cancel_order()
 ↓
FastAPI Backend APIs
 ↓
In-memory Data Store

---

Installation

Clone the repository:

git clone https://github.com/yourname/ai-agent-ecommerce-backend.git
cd ai-agent-ecommerce-backend

Create a virtual environment:

/opt/homebrew/opt/python@3.11/bin/python3.11 -m venv venv

Activate the environment:

Mac / Linux:

source venv/bin/activate

Windows:

venv\Scripts\activate

Install dependencies:

pip install -r requirements.txt

---

Install Ollama

Download and install Ollama from the official website:

https://ollama.com/download

After installation, start the Ollama server:

ollama serve

Using Cloud Models

This project can use Ollama cloud models such as:

• gpt-oss:20b-cloud
• deepseek-v3.1-cloud

Make sure you login first:

ollama login

Run Backend Server

Start the FastAPI server:

uvicorn backend.server:app --reload

Server will run at:

http://127.0.0.1:8000

API documentation:

http://127.0.0.1:8000/docs

---

Testing

Run agent evaluation tests:

pytest tests/test_agent.py -s

---

Learning Objectives

This project demonstrates:

• AI agent tool calling
• Backend API interaction
• FastAPI service development
• Local LLM integration
• AI + backend architecture design

---

Future Improvements

Potential extensions:

• PostgreSQL database
• Retrieval-Augmented Generation (RAG)
• Chat UI interface
• Authentication and user accounts
• Order tracking integration
• Customer support ticket system