This project is an AI-powered chatbot that allows users to query a supplier and product database using natural language.
It retrieves relevant product and supplier details from a MySQL database and summarizes them using Gemma-2-9b-it via the Groq API.
The chatbot is built using LangGraph for agent workflows and FastAPI for backend communication.
✅ Natural Language Querying – Users can ask about products, suppliers, and categories.
✅ LLM-Powered Summarization – Responses are enhanced using Gemma-2-9b-it.
✅ Real-Time Conversations – The chatbot maintains a session-based conversation history.
✅ Responsive UI – Built with React, Vite, and Tailwind CSS for a smooth experience.
✅ Efficient Database Queries – Uses MySQL for structured storage of products and suppliers.
🔹 Backend: Python, FastAPI, LangGraph, Groq API (Gemma-2-9b-it), MySQL
🔹 Frontend: React, Vite, Tailwind CSS, Axios
🔹 Database: MySQL
These are the tables(products, suppliers) of the database(ProductSuppliers.db) respectively:
, while
| ID | Name | Brand | Price | Category | Description | Supplier ID |
|---|---|---|---|---|---|---|
| 1 | Smartphone | X | 699.99 | Electronics | Latest model with AI camera | 1 |
| 2 | Refrigerator | Y | 1199.99 | Home Appliances | Energy-efficient cooling | 1 |
| 3 | Sofa Set | Z | 499.99 | Furniture | Comfortable 3-seater | 2 |
| 4 | Dining Table | Z | 899.99 | Furniture | Solid wood with six chairs | 2 |
| 5 | Organic Juice | W | 5.99 | Groceries | Freshly squeezed organic juice | 3 |
- The "products" table stores details about products, including brand, price, category, and descriptions.
| ID | Name | Contact Info | Product Categories |
|---|---|---|---|
| 1 | A | contactA@example.com | Electronics, Home Appliances |
| 2 | B | contactB@example.com | Furniture, Decor |
| 3 | C | contactC@example.com | Groceries, Beverages |
- The "suppliers" table holds supplier information such as contact details and product categories they provide.
(You only need to setup frontend and backend for testing purpose, rest are managed...).
- Python 3.9+
- MySQL installed & running
git clone https://github.com/your-username/chatbot-assessment.git
cd chatbot-assessment/backend
python -m venv venv
source venv/bin/activate # On Windows, use `venv\Scripts\activate`
pip install -r requirements.txt
python api.py
- node.js 18+ install from here
- vite
cd ../frontend
npm install
npm run dev
The app will be available at http://localhost:5173/ (backend will be running at http://localhost:8000/).
Try these sample queries:
❓Show me all the products.
❓name all products under brand X
❓Which suppliers provide furniture?
❓Give me details of product 2, 3 and 4.
❓Calculate the average price of all the available products.
❓which supplier should I contact for electronics? Also give me details for contacting them
❓which supplier is selling the most number of products? Also give me the names of all that products.
❓Which is the least priced product?