MongoDB Lab Assistant

The MongoDB Lab Assistant is a chatbot designed to assist users with questions about the MongoDB Lab Workshop. It uses OpenAI to generate embeddings for questions and answers from a markdown file and stores these embeddings in a MongoDB database. The chatbot then uses these embeddings to find and return relevant answers to user queries.

Features

• Question and Answer Embeddings: The system reads a markdown file containing questions and answers, generates embeddings for each, and stores them in MongoDB.
• Vector Search: Utilizes MongoDB's vector search capabilities to find relevant documents based on user queries.
• Chat Interface: Provides a chat interface for users to ask questions and receive answers based on the stored data.
• Debug Mode: Includes a debug mode to provide detailed information about the embeddings, relevant documents, and context used to generate responses.

Setup and Installation

Prerequisites

• Node.js
• MongoDB
• OpenAI API Key

Installation

1. Clone the repository:

   git clone https://github.com/your-repo/mongodb-lab-assistant.git
   cd mongodb-lab-assistant

2. Install dependencies:

   npm install

3. Set up environment variables: Create a .env file in the root directory and add the following environment variables:

   OPENAI_API_KEY=your_openai_api_key
   MONGODB_URI=mongodb+srv://your_mongo_user:your_mongo_password@your_cluster.mongodb.net/lab?retryWrites=true&w=majority&appName=performance
   MONGODB_DB=lab
   MONGODB_COLLECTION=documents
   REACT_APP_SERVER_URL=http://localhost:3001
   DEBUG_MODE=true

4. Generate Embeddings: Use the generate_embeddings.py script to read the markdown file and store the embeddings in MongoDB:

   python3 generate_embeddings.py

Running the Application

1. Start the server:

   npm run start:server

2. Start the client:

   npm start

Usage

Asking Questions

Users can interact with the chat interface to ask questions about the MongoDB Lab Workshop. The chatbot will search the database for relevant questions and answers and return the most appropriate responses.

Debug Mode

When DEBUG_MODE is set to true, additional debug information will be provided with each response, including the query embeddings, relevant documents, and context used.

Directory Structure

├── LICENSE
├── README.md
├── client
│ ├── Chatbot.css
│ ├── README.md
│ ├── package-lock.json
│ ├── package.json
│ ├── public
│ │ ├── favicon.ico
│ │ ├── index.html
│ │ ├── logo192.png
│ │ ├── logo512.png
│ │ ├── manifest.json
│ │ └── robots.txt
│ └── src
│ ├── App.css
│ ├── App.js
│ ├── App.test.js
│ ├── Chatbot.css
│ ├── index.css
│ ├── index.js
│ ├── logo.svg
│ ├── reportWebVitals.js
│ └── setupTests.js
├── docs
│ ├── generate_embeddings.py
│ ├── troubleshooting.md
├── package-lock.json
├── package.json
├── server
│ ├── chatbot.log
│ └── server.js
└── src
├── components
│ ├── BrowserWindow
│ │ ├── index.js
│ │ └── main.css
│ ├── Chatbot.js
│ ├── HomepageFeatures
│ │ ├── index.js
│ │ └── styles.module.css
│ ├── Link.js
│ ├── Screenshot.js
│ └── Sidebar.js
├── css
│ └── custom.css
└── pages
├── ChatPage.css
├── chat.js
├── helloWorld.js
├── index.js
└── index.module.css

Contributing

Contributions are welcome! Please open an issue or submit a pull request for any bugs or improvements.

License

This project is licensed under the MIT License - see the LICENSE file for details.