Coffee Shop Assistant ☕🧠

An intelligent, conversational assistant for coffee shop operations, powered by LangGraph. This agent helps users navigate the menu, place orders, and receive personalized suggestions—all through a multi-turn, context-aware chat interface. All chat history is persistently stored using a SQLite backend for seamless experiences across sessions.

---

🖼️ Application UI

The app features a clean, multilingual interface with real-time streaming chat, persistent session management, and interactive messaging components.

---

✨ Overview

Coffee Shop Assistant is designed to simulate a helpful AI agent in a real café setting. It guides customers through menu options, answers questions about drinks and items, and supports conversational follow-up using contextual memory.

The app architecture leverages:

• LangGraph for stateful, multi-step conversations.
• Streamlit for the interactive UI.
• SQLite for persistent chat storage.
• Gemini 2.5 Pro for fast, intelligent responses.
• Local database files for coffee menu data and interactions.

---

⚙️ How It Works

The assistant follows a structured LangGraph flow:

1. Initial Greeting: Welcomes the user and presents a helpful introduction.
2. Intent Recognition: Parses user intent—whether it’s asking about drinks, placing an order, or general questions.
3. Data Lookup: Fetches responses from internal knowledge (e.g. .db file) using the relevant LangGraph tools.
4. Response Generation: Crafts a helpful reply based on the query and current conversation context.
5. Persistence: Logs every message into a persistent SQLite conversation history.

---

🧠 Graph Architecture

The architecture relies on a deterministic graph built with LangGraph that controls user interactions in clear, logical steps. It includes memory nodes, conditional branching, and fallback logic for unrecognized intents.

---

🚀 Key Features

• Conversational Agent: Handles Arabic or English customer queries with natural language responses.
• Persistent Chat Memory: Saves each session in a local SQLite database.
• Interactive Streamlit UI: Streamlined, user-friendly experience.
• Menu Lookup: Supports intelligent retrieval of menu items and their details.

---

🛠️ Tech Stack

• Orchestration: LangChain & LangGraph
• LLM: Google Gemini (gemini-2.5-pro)
• Database: SQLite (coffee_shop.db)
• UI Framework: Streamlit
• Environment Management: Python venv or Docker

---

📦 Setup & Installation

Option 1: Run Locally (venv)

1. Clone the Repository

   git clone https://github.com/eslammohamedtolba/Coffee-Shop-Assistant.git
   cd Coffee-Shop-Assistant

2. Create and Activate a Virtual Environment

   # Create the environment
   python -m venv venv
   
   # Activate on Windows PowerShell
   .\venv\Scripts\Activate.ps1
   
   # On macOS/Linux
   source venv/bin/activate

3. Install Dependencies

   pip install -r requirements.txt

4. Add Environment Variables

   • Create a .env file in the root directory and add:

     GOOGLE_API_KEY="YOUR_GOOGLE_API_KEY"

5. Launch the App

   streamlit run main.py

6. Access the App
   Open http://localhost:8501 in your browser and start chatting.

---

Option 2: Run with Docker

1. Ensure Docker is Running

2. Build the Docker Image

   docker build -t coffee-shop-assistant .

3. Run the Container

   docker run -p 8501:8501 \
     -v "$(pwd)/coffee_shop.db:/app/coffee_shop.db" \
     -v "$(pwd)/.env:/app/.env" \
     coffee-shop-assistant

   Explanation:

   • -p 8501:8501: Maps Streamlit port
   • -v "$(pwd)/coffee_shop.db:/app/coffee_shop.db": Mounts your local database
   • -v "$(pwd)/.env:/app/.env": Injects your local API keys

4. Access the Application
   Go to: http://localhost:8501

5. Stop the Container

   docker stop coffee-shop-assistant

6. (Optional) Remove the Container

   docker rm coffee-shop-assistant

---

🤝 Contributing

Feel free to fork this repo, open issues, or submit pull requests to enhance functionality, add features, or improve performance.