How to get clean web data for chatbots and LLMs

This is a companion repository to go with František Nesveda's Dev World 2025 talk: How to get clean web data for chatbots and LLMs.

The presentation slides are available here in PDF and PPTX formats.

Chatbot examples

To run the chatbot examples, you need to have Node.js installed and install dependencies with:

npm install

To run the chatbots, you will need to export your OpenAI API key as an environment variable or use an alternative way of setting this env var:

export OPENAI_API_KEY=your-api-key

Finally, run them with:

node tesla-chatbot.js

node bmw-chatbot.js

For more information on how they work. Reference Crawlee and LangChain JS documentation.