Structuring the Unstructured
summary
Claims that a local open source pipeline converts PDFs, scans, tables and images into structured JSON/Markdown, demonstrates exporting eight tables from one paper and handling a 418‑section annual report, and shows Docling CLI and library plus a REST service and MCP server that actually run on CPU. Benchmarks report 50× cost savings on CPU versus naive VLM OCR. Demo builds chunkless retrieval via chunkless RAG and exports pydantic data that applications can consume.
AI-written summary · about
chapters (8)
- 0:00Unstructured data challenges
- 2:07Advanced document processing
- 5:57Docling for conversion
- 9:47Live PDF conversion demo
- 12:12Extracting images from docs
- 14:17Chunkless RAG with Docling
- 16:44Scaling Docling with REST API
- 19:38Summary and open-source
related talks
22mStructuring the Unstructured : Empowering the AI Workflows
26mAdvanced Document Parsing for AI Workflows
01h 19mLearn to Unlock Document Intelligence with Open-Source AI
01h 30mUnlock Information from Tables, Images and Complex Docs
32mDocuments Meet LLMs: Tales from the Trenches
38mBuild AI-Ready Search: Integrating Docling with OpenSearch for Advanced RA...
31mPDFs - When a thousand words are worth more than a picture (or table).
48mIntegrating Docling With OpenSearch for Advanced RAG and Agentic Applications