Databricks introduces a unified approach to Intelligent Document Processing (IDP) by combining Lakeflow and Document Intelligence. Lakeflow Connect provides zero-maintenance ingestion from SharePoint and Google Drive directly into Unity Catalog Volumes with built-in governance. Document Intelligence offers AI functions like ai_parse_document and ai_extract to parse, structure, and enrich complex documents (PDFs, scanned pages, handwritten notes) within existing data pipelines. Lakeflow Jobs orchestrates the full IDP workflow with advanced control flow, serverless compute, and real-time observability. Unity Catalog provides unified governance across structured and unstructured data, enabling enterprise-context-aware agentic workflows.
Table of contents
Step 1: Secure Ingestion with Lakeflow ConnectStep 2: Getting started with Databricks Document IntelligenceStep 3: Productionizing IDP Workloads at ScaleSort: