Planetary-scale answers, unlocked.
A Hands-On Guide for Working with Large-Scale Spatial Data. Learn more.
Authors
✨ Welcome to this month’s newsletter! Whether you’re comparing performance, exploring isochrones for spatial analysis, or implementing ML models for feature detection in satellite imagery, there’s something here for everyone. Read on for expert insights, product updates, and upcoming events you won’t want to miss!
🚨Last Call: Don’t miss your chance to try Wherobots Pro—free on AWS until May 31st!
If you’ve been meaning to explore Wherobots, now’s the perfect time. Get full access to the Pro tier, including unlimited performance and scalability for production workloads. Take advantage of powerful features like raster inference, map matching, and travel isochrones—only available in the Pro tier.
💡GeoPandas vs. Wherobots: Which is better for spatial analysis?
We ran a spatial join between the Overture Buildings dataset and a postal codes dataset.
Wherobots consistently outperforms GeoPandas in terms of speed—especially at scale. In fact, GeoPandas was unable to handle the largest dataset due to memory limitations.
That said, GeoPandas is great for smaller datasets. Its interoperability with lots of libraries makes it a great tool for lightweight spatial workflows.
Wherobots & Sedona are built for scale. It supports both Python and SQL APIs. And it’s ideal for large datasets and SQL-based workflows.
Combine these tools for the best of both worlds. Flexibility for small and large datasets with access to a broader ecosystem of libraries. Find out how Sedona and GeoPandas are better together.
How Wherobots Generated Drive Time Isochrones for Every POI in the US
📍Imagine if your geospatial models could reason in terms of real travel time rather than just distance.
⭐ Key highlights include:
Learn how Wherobots generated drive-time isochrones for every POI in the U.S., and how this data can transform planning in urban analytics, retail, transportation, and more.
Geospatial Tables in the Open Lakehouse: A New Era for Iceberg & Parquet
🧊With geospatial data now supported natively in Apache Iceberg and Parquet, spatial analytics becomes as scalable and performant as any other big data workload.
Join leaders from Planet, Databricks, Foursquare, and Wherobots in this insightful discussion as they explore the impact of native GEO support in Iceberg, including its core functionality, performance benefits, industry adoption, strategies for avoiding vendor lock-in, and what’s ahead for the future of geospatial data. 🔮
What if you could detect features in satellite imagery using natural language– without relying on GPU-intensive infrastructure?
With Raster Inference supporting Meta’s Segment Anything 2 (SAM2) model, you can now perform object detection and feature segmentation at scale on terabytes of satellite imagery—just by describing what you’re looking for in plain text.
Instead of downloading models or manually wrangling raster data, just run everything on Wherobots’ distributed cloud compute.
Query massive satellite datasets using natural language: “Find container ships” 🚢 or “Segment pickleball courts” 🎾—no GPUs, no model downloads, just cloud-native inference.
Because these results are stored as Iceberg tables in your S3 bucket, you can easily join with other datasets using WherobotsDB and continue your analysis inline with 300+ spatial features and functions.
Currently it supports OWLv2 SAM2, and you can go from prompt to polygons in a single SQL query.
🚀 Wherobots is now available in the AWS Europe (Ireland) region!
The EU leads in innovation across climate solutions, mobility, automotive design, precision agriculture, and urban development—industries that all rely heavily on geospatial data. Yet modern cloud tools haven’t kept pace, making geospatial development costly and complex, often requiring specialized teams and infrastructure.
For teams working in the EU, this launch offers lower latency support for European data residency requirements and simplified scaling—with no infrastructure headaches.
Cloud Native Geospatial Analytics with Apache Sedona (published by O’Reilly)
This latest chapter dives deeper intro the integration between Apache Sedona and the Python data science stack—empowering spatial workflows that scale using tools you already know and love.
If you’re following the series or want to catch up:
🔹 Chapter 1: Introduction to Apache Sedona
🔹 Chapter 2: Getting Started with Apache Sedona
🔹 Chapter 3: Working With Geospatial Data at Scale (Data Loading)
🔹 Chapter 4: Vector Data Analysis with Spatial SQL (Points, Lines, and Polygons)
💡 Latest: Sedona + PyData Ecosystem
📥 Get the hands-on guide for working with large-scale spatial data.
Apache Sedona Community Office Hour
Missed the most recent office hour? Watch the recording here. We covered:
📅 Mark your calendar for the next session, where we’ll cover upcoming Sedona 1.8.0 release:
Connect with us at the Data+AI Summit
Our team is heading to the Data+AI Summit in San Francisco, and we’d love to connect in person!
🎤 Don’t miss our session: Iceberg Geo Type: Transforming Geospatial Data Management at Scale, presented by Jia Yu (Wherobots) and Szehon Ho (Databricks)
📍 Find us at booth #E406 for free SWAG and to learn more about geospatial ETL.
🍻 Join us Wednesday evening at the Apache Iceberg Meetup, featuring Iceberg GEO and building mobility knowledge graphs with PuppyGraph.
📧 Book a meeting with us here to find a time to connect.
Join us at the EO Summit
If you’re heading to the EO Summit in New York, we’d love to learn about how you’re working with earth observation data. Share your use cases and let’s explore how Wherobots can help. 📧 Reach out at info@wherobots.com to schedule a time to meet!
Key takeaways
Links to GeoPandas vs Wherobots performance, the U.S. POI isochrones dataset, an Iceberg/Parquet GEO discussion, SAM 2 raster inference, the AWS Europe (Ireland) launch, Sedona community updates, and Data+AI Summit / EO Summit events.
Yes. It says Raster Inference supports Meta SAM 2 with OWLv2, so you can go from a natural-language prompt to polygons in a single SQL query on Wherobots distributed cloud compute, with results stored as Iceberg tables.
Yes. It announces availability in the AWS Europe (Ireland) region for lower latency and European data-residency requirements.
The next office hour was slated to cover the GeoPandas API on Sedona, Sedona on PyFlink, Spark 4.0 support, vectorized Python UDFs, and GeoParquet / Parquet Geo types.
RasterFlow is now available in Public Preview
RasterFlow makes planetary-scale earth intelligence workflows easy and costs predictable. We are excited to announce that RasterFlow is now in Public Preview, opening up the power of planetary scale Earth Intelligence to all Wherobots Professional Edition customers! RasterFlow let’s you solve complex monitoring challenges with vision-language models or tailored models for specific use cases, without […]
Wherobots for QGIS: Cloud-Scale Spatial Data, from Wherobots Labs
Wherobots for QGIS is a new plugin that connects QGIS to Wherobots Cloud. From a panel inside QGIS, an analyst can run spatial SQL against WherobotsDB, load the results as a map layer, push a local layer up to a Wherobots Iceberg table, and pull raster data into the canvas. Why Wherobots for QGIS: Eliminating […]
How Wherobots builds with NVIDIA to let AI see the physical world
The world and what happens in it is digitized by petabytes of raw and derivative spatial datasets of various data types and scales, and the potential for applying AI to it is immense. But the AI models and agents we use every day need connectivity to tools that turn this data into usable insights and relationships. Wherobots gives AI the ability operate on and understand raw physical-world data, and NVIDIA GPUs are core to it. Architecturally here’s how this works at a high level.
El Niño 2026, atmospheric rivers, and California’s burn scars: one SQL engine, two data models
7,713 mapped building footprints sit inside or within 500 m (1,640 ft) of the 42 Eaton basins USGS rates high hazard for debris flows, and 388 km (241 mi) of mapped road and path cross the low ground below the scar. WherobotsDB finds both in SQL: a join to the USGS basins for the buildings, and a raster vector join over elevation for the roads.
share this article
Awesome that you’d like to share our articles. Where would you like to share it to:
