☕ Layl Chatbot – AI-Powered Cafe Assistant

Welcome to the Layl Cafe Chatbot — an AI-driven customer support and conversational commerce platform. Built using an MVC architecture, it combines powerful LLMs, a multi-agent system, and Retrieval-Augmented Generation (RAG) to deliver a smart, interactive, and personalized user experience on a sleek, responsive React frontend.

The chatbot can handle complex conversational orders, answer nuanced menu questions, and provide intelligent product recommendations.

---

✨ Key Features

• Conversational Commerce: Users can place, modify, and inquire about their orders in natural language.
• Contextual RAG Q&A: Fetches accurate, up-to-date information from a Pinecone vector DB to answer specific questions about the cafe and its products.
• Smart Recommendations: Utilizes market basket analysis to suggest relevant, complementary items based on cart contents.
• Multi-Agent Workflow: A robust agent-based system intelligently routes user requests to the appropriate specialized tool.
• Full Admin Panel: A FastAPI-based interface allowing full CRUD (Create, Read, Update, Delete) operations for products, orders, and chat history.
• Persistent & Unified Sessions: Manages user state across the entire site (shopping cart, chatbot) using a centralized, cookie-based session system with MongoDB for data persistence.
• Advanced Text Chunking: Strategically groups related information for more effective document retrieval in the RAG pipeline.

---

🧠 Agent-Based Architecture

The chatbot uses modular agents, each with a dedicated role, collaborating in a pipeline:

Guard → Classification → Specialized Agents (Order / Details / Recommendation / Specialist)

Agent	Role
Guard Agent	Ensures user messages are on-topic and safe, blocking irrelevant or harmful queries.
Classification Agent	Acts as the central router, detecting user intent (e.g., ordering, asking a question) to activate the correct downstream agent.
Order Agent	Manages the entire conversational order workflow using chain-of-thought logic to collect and structure order details.
Details Agent	Analyzes the query to search the most relevant collection in the Pinecone VectorDB, using the retrieved chunks to generate a factual answer.
Recommendation Agent	Suggests complementary products based on the user's current context and order history.
Specialist Agent	Handles general conversational messages, such as greetings, goodbyes, and thank yous, for a more natural interaction.

---

🚀 Getting Started

Follow these instructions to set up and run the project locally.

Prerequisites

• Python 3.9+
• Node.js 18.x and npm/yarn
• Docker and Docker Compose
• Git

1. Clone the Repository

git clone [https://github.com/Mando-03/Mini-RAG-App.git](https://github.com/Mando-03/Mini-RAG-App.git)
cd Mini-RAG-App

2. Environment Setup

The backend requires API keys and a database connection string.

1. Navigate to the backend directory: cd backend
2. Create a new file named .env by copying the example file:

   cp .env.example .env

3. Open the .env file and add your credentials:

   MONGO_URI="your_mongodb_connection_string"
   PINECONE_API_KEY="your_pinecone_api_key"
   GEMINI_API_KEY="your_google_gemini_api_key"
   

3. Backend Setup

# From the backend/ directory
# Create and activate a virtual environment
python -m venv venv
source venv/bin/activate

# Install dependencies
pip install -r requirements.txt

# Run the FastAPI server
uvicorn main:app --reload --host 0.0.0.0 --port 5000

4. Frontend Setup

# Open a new terminal and navigate to the frontend/ directory
cd frontend

# Install dependencies
npm install

# Start the React development server
npm run dev

The application should now be running, with the frontend at http://localhost:3200 and the backend at http://localhost:5000.

---

🐳 Docker Deployment

For a production-like setup, you can build and run the entire application using Docker Compose.

1. Make sure your .env file is correctly configured in the backend/ directory.
2. From the project's root directory, run:

   docker-compose up --build

This will build the images for both the frontend and backend services and run them together.

---

🖥️ Frontend – React Cafe Website

A modern, user-friendly web interface that integrates directly with the chatbot.

⚙️ Built With

• React 18 – Fast, responsive UI
• Vite – Ultra-fast dev server & build tool
• TailwindCSS – Utility-first styling
• Redux Toolkit – State management
• React Router v6 – Routing

---

🧰 Backend & AI Stack

• FastAPI, Uvicorn – Asynchronous Backend API
• MongoDB + Motor – Async NoSQL Database
• Pinecone – Vector Database for RAG
• SentenceTransformers, Gemini, LangChain – NLP & LLMs
• Docker – Containerization and Deployment