Integrating Docling With OpenSearch for Advanced RAG and Agentic Applications
summary
Docling converts PDFs, DOCX, PPTX and images into a unified Docling document that preserves layout, tables, figures and metadata for RAG; it ships a hybrid chunker that chunks by document hierarchy (sections, headings, tables) and generates embeddings for OpenSearch vector indexes; integrations include Docling Surf (REST/UI/CLI), a Java API and an OpenSearch ingest plugin in progress; chart extraction and XBRL support are available, and GPU OCR demos with Nvidia report 4–12 pages per second on modern GPUs.
AI-written summary · about
chapters (12)
- 0:00Docling and OpenSearch for RAG
- 4:38Document extraction challenges
- 10:37Docling as document processor
- 13:30Live demo: Docling extraction
- 14:58Docling pipelines and architecture
- 20:20Enriching data for AI
- 21:23Chunking strategies with Docling
- 23:29Docling with OpenSearch embeddings
- 26:38Docling agentic apps
- 29:28Open RAG platform
- 30:35Docling community and future
- 35:57Q&A
related talks
38mBuild AI-Ready Search: Integrating Docling with OpenSearch for Advanced RA...
01h 00mComplex Data Ingestion with Open Source AI
01h 19mLearn to Unlock Document Intelligence with Open-Source AI
27mWhy Your RAG is Failing , And How Open Source Can Fix It
16mOpenRAG: An open-source stack for RAG
01h 30mUnlock Information from Tables, Images and Complex Docs
26mAdvanced Document Parsing for AI Workflows
34mImplementing an open-source RAG with OpenSearch