Turn ANY File into LLM Knowledge in SECONDS
36,989 views Oct 1, 2025
One of the biggest challenges we face with LLMs is their knowledge is too general and limited for anything new. That’s why RAG is so popular - it’s a method for providing an LLM with external knowledge you curate so it can become an expert on your data. The problem is, that “curate” step can be very difficult if you have data in a lot of different formats.
That is where Docling comes in! Docling is an open source data pipeline and chunking framework specifically designed to handle all your data formats and prepare them for LLMs.
In this video, I show you how to use Docling to extract text from virtually ANY file type and chunk it perfectly for a RAG system. Plus at the end, I even show you a RAG AI agent I built that uses Docling for the RAG engine which you can use as a template right now (link below)!
If you want to see Docling in action in a production ready RAG pipeline and AI agent, check out Dynamous (Docling workshop this Friday!):
https://dynamous.ai
Docling RAG Agent and examples:
https://github.com/coleam00/ottomator-agents/tree/main/docling-rag-agent
Docling GitHub repository:
https://github.com/docling-project/docling
Docling documentation:
https://docling-project.github.io/docling/
00:00 - Introducing Docling - RAG Made Easy
01:36 - Getting Started with Docling
03:33 - Dynamous Event - Full RAG Pipeline with Docling
04:04 - Example #1 - PDF Parsing
06:26 - Example #2 - Working with Different File Types
08:24 - Example #3 - Extract Text from Audio Files
10:30 - Example #4 - Hybrid Chunking
14:26 - More Docling Resources (So Many Examples!)
15:41 - Grand Finale - Docling RAG AI Agent
20:36 - Final Thoughts