BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//talks.osgeo.org//foss4g-europe-2026//talk//JM9A8T
BEGIN:VTIMEZONE
TZID:EET
BEGIN:STANDARD
DTSTART:20001029T050000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:EET
TZOFFSETFROM:+0300
TZOFFSETTO:+0200
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000326T040000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:EEST
TZOFFSETFROM:+0200
TZOFFSETTO:+0300
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-foss4g-europe-2026-JM9A8T@talks.osgeo.org
DTSTART;TZID=EET:20260701T153000
DTEND;TZID=EET:20260701T160000
DESCRIPTION:Large Language Models are reshaping how we interact with data 
 — but most implementations ignore geography entirely. At Camptocamp\, we
 've spent the last two years embedding LLMs deep into open source geospati
 al workflows\, and this talk is a frank account of what works\, what doesn
 't\, and where the field is heading.\n\n** GeoNetwork as a GeoAI laborator
 y\n\nGeoNetwork\, the OSGeo flagship metadata catalog\, is where much of o
 ur work has been grounded. We'll walk through the integration of semantic 
 search — moving beyond keyword matching to meaning-based retrieval power
 ed by embedding models — and the development of a conversational assista
 nt that lets users query geographic datasets in plain language. We'll also
  share our ongoing work on exposing GeoNetwork capabilities through the Mo
 del Context Protocol (MCP)\, enabling LLM agents to interact directly with
  catalog APIs.\n\n*** Agentic geospatial: bleeding edge techniques\n\nBeyo
 nd search and chat\, we'll dive into what agentic AI looks like when appli
 ed to geospatial workflows: function calling to orchestrate GIS operations
  (buffer\, intersection\, spatial queries against OpenStreetMap)\, LLM-dri
 ven QGIS automation via MCP\, and the architectural patterns — RAG pipel
 ines\, intent extraction\, hybrid search — that make these systems relia
 ble enough to put in front of real users.\n\n** The French National Digita
 l Twin: an open source GeoAI at scale\n\nWe'll close with our role leading
  the LLM workstream of the French National Digital Twin project (France 20
 30)\, a consortium bringing together IGN\, INRIA\, Cerema and others. This
  initiative is tackling GeoAI at territorial scale — and doing it entire
 ly in the open. We'll share early architectural decisions\, the challenges
  of grounding LLMs in authoritative geographic knowledge bases\, and why o
 pen source is not just a preference here but a sovereignty requirement.\n\
 n** Key takeaways for the FOSS4G community\n\nAttendees will leave with a 
 clear picture of the current state of open source GeoLLM tooling\, practic
 al patterns for integrating LLMs into OSGeo-stack applications\, and an ho
 nest assessment of the remaining challenges — from data quality to model
  size optimization — that the community needs to solve together.
DTSTAMP:20260605T033444Z
LOCATION:A12
SUMMARY:GeoLLM in the Wild: Open Source AI Meets Geospatial - Florent Gravi
 n
URL:https://talks.osgeo.org/foss4g-europe-2026/talk/JM9A8T/
END:VEVENT
END:VCALENDAR