🍕 Pizza Restaurant AI Assistant

This is an AI-powered assistant for a pizza restaurant that supports:

• Conversational interaction via the command line
• A web server with a FastAPI backend
• Tool usage like placing pizza orders and querying documents

🚀 Features

• 🔧 Tool-based interactions with LLM agents
• 🍕 Place pizza orders by providing pizza type, size, quantity, and delivery address
• 📄 Query indexed documents (e.g., reviews, orders) using semantic search
• 🧠 Persistent conversation history
• 💾 Vector store support with ChromaDB and Ollama embeddings
• 🌐 REST API and WebSocket support via FastAPI

---

⚙️ Installation

1. Install and Setup Ollama

   First, install Ollama:

   curl -fsSL https://ollama.com/install.sh | sh

   Start the Ollama server in the background:

   nohup ollama serve &

   Download the required models:

   ollama pull llama3.2
   ollama pull mxbai-embed-large
   ollama pull llama3.2-vision

   Note: The Ollama server must be running for the assistant to function properly.

2. Clone the repo

   git clone https://github.com/patrickwide/PizzaAssist.git
   cd PizzaAssist

3. Create and activate a virtual environment

   python -m venv venv
   source venv/bin/activate  # For Linux/macOS
   .\venv\Scripts\activate   # For Windows

4. Install dependencies

   pip install -r requirements.txt

5. Initialize Data Structure

   ./create_data_structure.sh

   You should see the following output:

   🔧 Creating base directory structure...
   ✅ Created directory: data/db
   ✅ Created directory: data/history
   ✅ Created directory: data/documents
   📄 Creating placeholder files in data/documents...
   ✅ Created file: data/documents/orders.txt
   ✅ Created file: data/documents/realistic_restaurant_reviews.csv
   📝 Creating system_message.md and welcome_message.md...
   ✅ Created: data/system_message.md
   ✅ Created: data/welcome_message.md
   
   🎉 Directory and file structure successfully initialized under 'data/'
   ✅ Setup complete. You can start working with your project.
   

---

🎮 Test in Google Colab

You can easily test the Pizza Restaurant AI Assistant using Google Colab without any local setup:

The notebook provides an interactive environment to:

• Test the assistant's capabilities
• Place pizza orders
• Query documents
• Experiment with different prompts

Simply click the "Open in Colab" button above to get started!

---

🧠 Run in CLI Mode

This launches the assistant in the terminal with natural language support.

python cli.py

You'll see:

Welcome to the Pizza Restaurant Assistant!
Ask about reviews or place an order.
(e.g., 'How is the pepperoni pizza?', 'Tell me about the service',
       'I want to order 1 large veggie pizza to 456 Oak Avenue', 'exit' to quit)

---

🌐 Run as a Web Server

This launches the FastAPI backend on port 8000.

python main.py

You'll see logs like:

🚀 Server starting on http://127.0.0.1:8000
🔌 WebSocket endpoint: ws://127.0.0.1:8000/ws/ai
❤️  Health check: http://127.0.0.1:8000/health

API Endpoints

Method	Endpoint	Description
GET	/health	Health check
WS	/ws/ai	WebSocket for AI chat

---

📦 Run and Distribute as a Python Package

How to Install and Run Locally as a Package

1. Install the package in development mode

   pip install -e .

2. Run the web server

   pizza-assist-server

How to Build the Package

1. Build the distribution packages

   python -m build

   This will create both wheel (.whl) and source distribution (.tar.gz) files in the dist/ directory.

How to Distribute and Run in Production

1. Install from a distribution file

   pip install dist/pizza_assist-0.1.0-py3-none-any.whl

2. Run in production

   # Start the web server
   pizza-assist-server --host 0.0.0.0 --port 8000

---

🧪 Tooling

Available Tools

• place_pizza_order

  • Required: pizza_type, size, quantity, delivery_address
  • Description: Places a pizza order and saves it

• query_documents

  • Required: query
  • Description: Searches and retrieves content from indexed documents

---

📝 Logs

Logs are written to the logs/ directory for debugging and tracing interactions.

---

🧠 Memory & Vector Store

• Conversation History: stored in core/data/history/conversation_history.jsonl
• Vector Store: ChromaDB-powered local vector DB with Ollama embeddings

---

📄 License

MIT License

---

👨‍💻 Author

Made by patrickwide