Pisek K.’s Post

Ever feel like your RAG system is missing half the story by ignoring the images in your PDFs? It's a common problem that leads to lost information and context. I've put together a new video walking through how to solve this with a multi-modal RAG pipeline! I demonstrate how to use Docling, a powerful open-source tool, to: - Process PDFs containing both text and images. - Use AI to automatically describe images, creating what I call "enriched text". - Build a complete indexing pipeline for a vector database. The result is a more powerful and accurate agentic AI that can reason and select the right knowledge base to answer your questions. This is a great way to build advanced AI applications for your team. See the video and open-source code in the comments below. Let me know what you think! #MultiModalRAG #AI #PDFProcessing #Docling #RAG #AgenticAI