BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.devconf.info//devconf-us-2025//talk//XLZMFE
BEGIN:VTIMEZONE
TZID:EST
BEGIN:STANDARD
DTSTART:20001029T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10;UNTIL=20061029T070000Z
TZNAME:EST
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
END:STANDARD
BEGIN:STANDARD
DTSTART:20071104T030000
RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11
TZNAME:EST
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000402T030000
RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=4;UNTIL=20060402T080000Z
TZNAME:EDT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
END:DAYLIGHT
BEGIN:DAYLIGHT
DTSTART:20070311T030000
RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3
TZNAME:EDT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-devconf-us-2025-XLZMFE@pretalx.devconf.info
DTSTART;TZID=EST:20250920T124000
DTEND;TZID=EST:20250920T140000
DESCRIPTION:Most real-world data remains trapped in complex documents: PDFs
  with intricate layouts\, PowerPoints with embedded diagrams\, Word docume
 nts with nested tables. Traditional extraction tools fail extract valuable
  information from these documents that improve your AI workflows. Tables b
 ecome jumbled text\, figures disappear\, and document structure is lost. T
 his workshop introduces Docling\, an open-source toolkit that uses deep le
 arning to understand documents the way humans do.\n\nThrough three hands-o
 n labs\, you'll build a complete document processing pipeline:\nLab 1: Con
 vert complex documents (PDF\, DOCX\, PPTX\, HTML) into structured data whi
 le preserving tables\, figures\, and layouts. See how Docling maintains re
 lationships that other tools destroy.\nLab 2: Implement intelligent chunki
 ng strategies that respect document structure—critical for accurate retr
 ieval in AI applications.\nLab 3: Build a multimodal RAG system with visua
 l grounding\, a unique Docling feature that shows users exactly where info
 rmation originates in source documents.
\n\nYou'll leave with working co
 de for document processing pipelines and the skills to integrate Docling i
 nto your AI workflows. All processing runs locally on standard hardware.\n
 \nPrerequisites: Python 3.10+\, basic Python knowledge 
Target Audience:
  Developers and data scientists working with document-heavy workflows
DTSTAMP:20260310T065853Z
LOCATION:107 (Capacity 20)
SUMMARY:Mastering Multi-Format Document Processing for AI with Docling - Mi
 ngxuan Zhao\, Rafael Vasquez
URL:https://pretalx.devconf.info/devconf-us-2025/talk/XLZMFE/
END:VEVENT
END:VCALENDAR