How do you visualize decades of high-resolution national climate projection data — wind speed, solar radiation, temperature, degree days — directly in a web browser, with smooth rendering and precise data picking, all built on open source tools?

At Camptocamp, we tackled this challenge during the Météo France hackathon in Toulouse, where teams were given access to beta climate projection datasets from Météo France and the DINUM. Our goal: build a web application helping the renewable energy industry assess regional potential under different climate warming scenarios, aligned with the French government's TRACC adaptation framework (+2°C, +2.7°C, +4°C milestones).

The data pipeline — built with Python, GDAL, Xarray, and RioXarray — transforms NetCDF climate model outputs (CNRM-ALADIN64E1, ~12 km resolution, 2014–2100) into monthly raster tilesets. The challenge then becomes how to render these rasters with maximum expressiveness on a map, going well beyond what standard raster layer support in MapLibre GL JS offers out of the box.

This is where our open source library maplibre-gl-shader-layer comes in. Born from real production needs in meteorological data visualization, this TypeScript/WebGL library provides the building blocks to create fully custom tiled layers for MapLibre GL JS, powered by Three.js under the hood. Developers can write their own GLSL fragment shaders and hook into per-tile uniform updates — giving full control over color mapping, encoding, blending, and animation.

The library's flagship component, MultiChannelSeriesTiledLayer, is designed specifically for scientific data: it decodes multi-channel PNG or WebP tiles where RGB channels encode up to 24-bit precision float values (similar to Mapbox Terrain-RGB), supports time/depth/scenario series interpolation, and applies configurable colormaps from a built-in library (viridis, inferno, turbo, and custom descriptions). Nodata handling via the alpha channel and support for PMTiles archives round out the feature set for production use.

We will walk through the full open source stack — from raw NetCDF to interactive browser map — and show how maplibre-gl-shader-layer makes it straightforward to build expressive, performant meteorological visualizations without sacrificing flexibility. Demos will include climate indicator overlays, a warming scenario slider, and seasonal navigation — all rendered in WebGL.

The library is MIT-licensed, available on npm, and actively maintained by the Camptocamp geoblocks team.

Repository: https://github.com/geoblocks/maplibre-gl-shader-layer

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 geospatial workflows, and this talk is a frank account of what works, what doesn't, and where the field is heading.

** GeoNetwork as a GeoAI laboratory

GeoNetwork, the OSGeo flagship metadata catalog, is where much of our work has been grounded. We'll walk through the integration of semantic search — moving beyond keyword matching to meaning-based retrieval powered by embedding models — and the development of a conversational assistant that lets users query geographic datasets in plain language. We'll also share our ongoing work on exposing GeoNetwork capabilities through the Model Context Protocol (MCP), enabling LLM agents to interact directly with catalog APIs.

*** Agentic geospatial: bleeding edge techniques

Beyond search and chat, we'll dive into what agentic AI looks like when applied to geospatial workflows: function calling to orchestrate GIS operations (buffer, intersection, spatial queries against OpenStreetMap), LLM-driven QGIS automation via MCP, and the architectural patterns — RAG pipelines, intent extraction, hybrid search — that make these systems reliable enough to put in front of real users.

** The French National Digital Twin: an open source GeoAI at scale

We'll close with our role leading the LLM workstream of the French National Digital Twin project (France 2030), a consortium bringing together IGN, INRIA, Cerema and others. This initiative is tackling GeoAI at territorial scale — and doing it entirely in the open. We'll share early architectural decisions, the challenges of grounding LLMs in authoritative geographic knowledge bases, and why open source is not just a preference here but a sovereignty requirement.

** Key takeaways for the FOSS4G community

Attendees will leave with a clear picture of the current state of open source GeoLLM tooling, practical patterns for integrating LLMs into OSGeo-stack applications, and an honest assessment of the remaining challenges — from data quality to model size optimization — that the community needs to solve together.