GeoNYC: March 2026 Edition

​Whether you missed the last GeoNYC event or had a great time and want more, we’re excited to be back with another one, co-hosted with AWS.

​GeoNYC brings together practitioners, technologists, researchers, and anyone curious about geospatial data. From building modern geospatial tech stacks to scaling AI-ready Earth observation pipelines, we’ll explore how organizations are using spatial data to solve real business challenges across industries.

​AGENDA

Fast, interactive raster data visualization the browser with deck.gl-raster | Kyle Barron from Development Seed

​We're creating a new ecosystem for client-side raster data visualization in the browser, enabling interactive WebGL rendering of COG data.

​This talk presents a high-level overview of how this works and how to leverage it in your projects.

GeoZarr and Icechunk for Planetary-Scale Rasters and Embeddings | Ryan Abernathey from Earthmover

Zarr is a community-driven cloud-native format for storing massive-scale n-dimensional arrays (tensors) which has taken off in the weather and climate space. This talk will review recent developments in the Zarr ecosystem which make it more useful for Earth-observation geospatial applications, specifically:
- The GeoZarr conventions, which provide a straightforward specification for encoding geospatial metadata (CRS, coordinates, and overviews) using standard Zarr attributes
- The Icechunk storage engine, which provides ACID transactions and data versioning, turning Zarr into a scalable array database

​Learn how these tools are being used in cutting-edge application such as Sentinel EOPF (next-generation Sentinel format) and in new global-scale GeoAI embedding datasets such as Tessera.

​Accelerating Spatial Analytics with Wherobots Spatial AI Coding Assistant | Kyle Ryan from Wherobots

​Geospatial data ecosystems are growing faster than any team can keep up with. This session walks through how the Wherobots Spatial AI Coding Assistant turns that complexity into a conversation, letting developers interrogate live catalogs, run multi-dataset spatial joins, and ship production jobs without ever leaving VS Code.

​You will see real validated queries, a working infrastructure vulnerability assessment built entirely through natural language, and a clear picture of what it looks like when one practitioner can do the work of five.