An intelligent chatbot application designed to help developers with OpenTelemetry integration and instrumentation. Built with Node.js, React, and powered by multiple LLM providers through LangChain with RAG (Retrieval Augmented Generation) capabilities.

• Multi-LLM Support: Compatible with OpenAI, Anthropic Claude, and AWS Bedrock
• RAG-Powered Responses: Uses vector search to provide contextually relevant answers
• OpenTelemetry Expertise: Pre-loaded with comprehensive OpenTelemetry documentation
• Modern Web Interface: Clean, responsive React-based chat interface
• Real-time Streaming: Fast response generation with typing indicators
• Source Attribution: Shows which documents were used to generate responses
• Provider Switching: Easily switch between different AI providers

┌─────────────────┐    ┌─────────────────┐    ┌─────────────────┐
│   React Client  │────│  Express API    │────│   LLM Provider  │
│                 │    │                 │    │ (OpenAI/Claude/ │
│  - Chat UI      │    │  - Chat Routes  │    │     Bedrock)    │
│  - Provider     │    │  - Admin Routes │    └─────────────────┘
│    Selection    │    │  - Middleware   │    
└─────────────────┘    └─────────────────┘    
                                │              
                       ┌───────────────────┐    
                       │  Vector Store     │    
                       │   (ChromaDB)      │    
                       │                   │    
                       │  - OTel Docs      │    
                       │  - Code Examples  │    
                       │  - Best Practices │    
                       └───────────────────┘    

• Node.js 18+
• npm or yarn
• One or more LLM provider API keys:
  • OpenAI API key (mandatory)
  • Anthropic API key (optional)
  • AWS credentials (for Bedrock, optional)

1. Clone the repository

   git clone <repository-url>
   cd hny-ai-workshop

2. Run quick start script

   scripts/quick-start.sh

   The quick start script is a one-for-all script that checks and installs all the required libraries and starts up the whole thing. Here are the layout of the individual steps that you can perform individually:

   NOTE: When running quick-start.sh for the first time, during the startup of chromaDB, the script may time out and exit. When that happens, no need to worry. You can re-run the script and the script will continue from where it left off.

   OPTIONAL

   1. Install dependencies

      # install concurrently
      npm install --save-dev concurrently
      
      # Install server dependencies
      npm install
      
      # Install client dependencies
      cd client && npm install && cd ..
      
      # Or use the convenience script
      npm run install-all

   2. Configure environment variables

      cp env.example .env

      Edit .env with your configuration:

      # Choose your default LLM provider
      DEFAULT_LLM_PROVIDER=openai
      
      # Add your API keys
      OPENAI_API_KEY=your_openai_api_key_here
      ANTHROPIC_API_KEY=your_anthropic_api_key_here
      
      # AWS Bedrock (optional)
      AWS_ACCESS_KEY_ID=your_aws_access_key
      AWS_SECRET_ACCESS_KEY=your_aws_secret_key
      AWS_REGION=us-east-1

   3. Set up ChromaDB (Vector Database)

      Install and start ChromaDB:

      pip install chromadb
      chroma run --host localhost --port 8000

   4. Ingest OpenTelemetry documentation

      npm run setup-data

   5. Start the application

      # Development mode (starts both server and client)
      npm run dev
      
      # Or start server only
      npm start
      
      # Or start everything (client, server, and chromaDB)
      npm run start:all
      or
      scripts/quick-start.sh

3. Access the application

   • Frontend: http://localhost:3000
   • API: http://localhost:3001/api
   • Health check: http://localhost:3001/api/health

4. **Stop the application

   npm run stop:all

1. Open the web application in your browser
2. Select your preferred LLM provider from the dropdown
3. Ask questions about OpenTelemetry integration:
   • "How do I set up auto-instrumentation for Express?"
   • "What's the difference between manual and automatic instrumentation?"
   • "How can I create custom spans?"
   • "How can I instrument react web?"

• POST /api/chat - Send a chat message
• GET /api/chat/context - Get context for a question
• GET /api/chat/providers - List available providers
• POST /api/chat/test-provider - Test a provider

• POST /api/admin/ingest - Add documents to knowledge base
• GET /api/admin/vector-store/info - Get vector store info
• POST /api/admin/search - Search documents
• DELETE /api/admin/vector-store - Reset knowledge base

You can add your own documentation to the knowledge base:

curl -X POST http://localhost:3001/api/admin/ingest \
  -H "Content-Type: application/json" \
  -d '{
    "title": "Custom OTel Guide",
    "content": "Your documentation content here...",
    "source": "internal-docs",
    "metadata": {
      "type": "guide",
      "version": "1.0"
    }
  }'

hny-ai-workshop/
├── server/                 # Backend Express.js application
│   ├── config/            # Configuration and logging
│   ├── middleware/        # Express middleware
│   ├── routes/           # API route handlers
│   ├── services/         # Business logic services
│   └── index.js          # Server entry point
├── client/               # React frontend application
│   ├── src/
│   │   ├── components/   # React components
│   │   ├── services/     # API service layer
│   │   └── App.js        # Main app component
│   └── public/           # Static assets
├── scripts/              # Utility scripts
│   └── ingest-data.js    # Data ingestion script
├── data/                 # Data storage directory
└── docs/                 # Documentation

1. Install the LangChain integration package
2. Add provider configuration to server/config/index.js
3. Initialize the provider in server/services/llmProvider.js
4. Update environment variable documentation

npm test

# Build the React client
npm run build

# Start in production mode
NODE_ENV=production npm start

1. No traces/providers available

   • Check your API keys in .env
   • Verify the provider is properly configured
   • Check server logs for initialization errors

2. ChromaDB connection issues

   • Ensure ChromaDB is running on localhost:8000
   • Check if the collection exists
   • Try resetting the vector store

3. Frontend can't reach API

   • Verify the proxy configuration in client/package.json
   • Check that the backend is running on port 3001
   • Look for CORS issues in browser console

Enable debug logging:

NODE_ENV=development npm start

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Add tests if applicable
5. Submit a pull request

MIT License - see LICENSE file for details

• OpenTelemetry community for excellent documentation
• LangChain for RAG capabilities
• All the AI providers for making this possible

For more information, check out the OpenTelemetry documentation or ask the chatbot! 🚀