Wherobots Platform Overview

Wherobots has two products: WherobotsDB and RasterFlow. You use both in Wherobots Cloud, the managed service and web console where your Organization runs workloads, connects data, and manages members and billing. Wherobots was founded by the creators of Apache Sedona.

Ready to get started? Start a free Professional Edition trial and run your first notebook in minutes.

Core products

WherobotsDB

Lakehouse compute engine for spatial dataRuns Spatial SQL, Python, and Scala on Apache Iceberg tables and files in object storage, with 300+ vector and raster functions. Compatible with the Apache Sedona APIs.Modules:

GeoStats: DBSCAN, Getis-Ord Gi*, Local Outlier Factor
Map matching: GPS-to-road-network alignment
Vector tiles: PMTiles generation from vector data

RasterFlow

AI perception engine for Earth imagery Public PreviewBuilds mosaics from satellite and aerial imagery, runs computer vision models over them, and vectorizes the predictions for analysis in WherobotsDB.

Pre-configured models for land cover, roads, canopy height
Bring your own data and models
Multiscales for interactive viewing

WherobotsDB in depth

WherobotsDB is the Wherobots lakehouse compute engine for spatial data. It runs Spatial SQL, Python, and Scala workloads on Apache Iceberg tables and files in object storage, with 300+ vector and raster functions and Apache Sedona API compatibility. GeoStats, map matching, and vector tile generation are WherobotsDB modules.

Key capabilities

Unified vector and raster: Derive insights through spatial relationships between vector and raster data types in a single engine.
Apache Sedona compatible: Run existing Apache Sedona code with the same APIs.
Popular languages: Spatial SQL, Python, and Scala. No esoteric tools required.
Open formats: Apache Iceberg tables, Parquet, and GeoParquet, in your own cloud storage or in Managed Storage.
Feature complete: 300+ vector and raster functions, notebooks for Exploratory Data Analysis (EDA), production-ready Job Runs, and Data Catalogs for browsing catalogs.
Serverless and secure: No infrastructure management. 99.5% runtime SLA with isolation by default for Enterprise Edition Organizations.

Common use cases

Automating spatial data processing: Schedule and automate ETL with Job Runs. Orchestrate pipelines with the Wherobots Airflow Provider.
Modernizing Spark + Sedona workloads: Move existing Apache Sedona on Spark workloads to WherobotsDB without managing clusters.
Analyzing imagery predictions: Join the vectorized output of RasterFlow with your other spatial data.

Getting started

The fastest way to start is to run an example notebook:

Start a free Professional Edition trial.
Start a runtime and open examples/Getting_Started/Part_1_Loading_Data.ipynb.

This notebook introduces the catalogs in Data Catalogs, Spatial SQL, and geospatial visualization.Already have existing workloads? Use WherobotsRunOperator to run them as Job Runs from Airflow.

WherobotsDB modules

GeoStats, map matching, and vector tile generation are WherobotsDB modules. You call them from the same SedonaContext session as the rest of your WherobotsDB code.

GeoStats

Distributed spatial statistics and clustering algorithms for detecting hotspots, density patterns, and local outliers in vector data.Use cases: pedestrian activity hotspots, public health outbreak detection, strategic retail placement.Algorithms: DBSCAN, Getis-Ord Gi*, Local Outlier Factor.

sedona = SedonaContext.create(config)
clusters_df = dbscan(df, 0.3, 10, include_outliers=True)

GeoStats is available in all Organization Editions. See the GeoStats reference.

Map Matching

Offline distributed map matching — precisely align GPS or location-tracking coordinates to digital road networks.Use cases: traffic pattern analysis, optimized routing.

sedona = SedonaContext.create(config)
df_edge = matcher.load_osm("data/osm.xml", "[car]")
result_df = matcher.match(df_edge, df_paths, "geometry", "geometry")

Map Matching is available in all Organization Editions. See the map matching tutorial.

Vector Tiles

Generate vector tiles in PMTiles format from large vector datasets, for use in web maps, mobile apps, and desktop GIS software.

from wherobots import vtiles
vtiles.generate_pmtiles(features_df, "s3://my-bucket/tiles/features.pmtiles")

See the Tile Generation tutorial and the vtiles reference.

Edition availability

	Capability	Professional	Innovation	Enterprise
	GeoStats (DBSCAN, LOF, Gi*)	Yes	Yes	Yes
	Map Matching	Yes	Yes	Yes

Example notebooks:

Capability	Notebook Path
DBSCAN	`examples/Analyzing_Data/Clustering_DBSCAN.ipynb`
Local Outlier Factor	`examples/Analyzing_Data/Local_Outlier_Factor.ipynb`
Getis-Ord Gi*	`examples/Analyzing_Data/Getis_Ord_Gi*.ipynb`
Map Matching	`examples/Analyzing_Data/GPS_Map_Matching.ipynb`

Table storage

WherobotsDB stores tables in Apache Iceberg. New tables in Wherobots-managed catalogs, such as org_catalog, are format-version 3 (V3), which has geometry and geography column types, and Wherobots adds a raster column type. Tables created before September 15, 2026 were created at format-version 2 (V2), and Wherobots doesn’t raise them to V3. Which column types a table can have, and whether WherobotsDB skips data files for spatial filters, depend on its format version; see the Iceberg Tables Overview. Wherobots also adds the CREATE SPATIAL INDEX statement, which rewrites a table’s data files clustered by a geometry or raster column. For details, see Geometry and Geography Tables, Raster Tables, and Spatial Filter Push-down and Clustering. To check whether an existing table’s geometry columns have a declared CRS, and to copy a table whose columns don’t, see Upgrade Existing Geometry Tables.

RasterFlow in depth

RasterFlow is the Wherobots AI perception engine for Earth imagery. It builds mosaics from satellite and aerial imagery, runs computer vision models over them at scale, and vectorizes the predictions for analysis in WherobotsDB. RasterFlow is available to every Organization. To get started, see Get Started with RasterFlow.

How the products work together

WherobotsDB and RasterFlow can be used on their own or together. Here are some common patterns that combine the products and the WherobotsDB modules to build spatial data products:

Spatial ETL Pipeline

WherobotsDB + AirflowIngest raw spatial data from S3 or public sources, transform it with Spatial SQL, and store the results in Apache Iceberg tables. Automate your ETL with the Wherobots Airflow Provider.Typical flow: Data Source → WherobotsDB (transform) → Iceberg tables (store) → Downstream consumers

Satellite Imagery Analysis

RasterFlow + WherobotsDBUse RasterFlow to build mosaics from satellite imagery, run ML inference (classification, segmentation, object detection) with pre-configured or custom models, and vectorize the results. Then analyze and enrich the output with WherobotsDB.Typical flow: Satellite imagery → RasterFlow (mosaic + infer + vectorize) → WherobotsDB (analyze vectors) → Visualization

Location Intelligence

WherobotsDB with the GeoStats moduleLoad point-of-interest or movement data, enrich with spatial joins, and detect clusters, hotspots, or anomalies using GeoStats algorithms.Typical flow: Vector data → WherobotsDB (join + enrich) → GeoStats (cluster) → Insights

Traffic & Routing Analysis

WherobotsDB with the map matching moduleLoad GPS traces, match them to road network geometries, and analyze traffic patterns, congestion, or optimal routes.Typical flow: GPS data + road network → Map Matching → WherobotsDB (analyze) → Routing decisions

Key security properties

Wherobots is designed with security as a top priority. Wherobots has the SOC 2 Type 2 attestation. For the full details, see the Security Guide and Wherobots Trust Center.

Wherobots Cloud

Wherobots Cloud is the managed service and web console where you use WherobotsDB and RasterFlow. In it, you start runtimes for notebooks, Job Runs, and SQL Sessions, connect storage and browse catalogs in Data Catalogs, and manage your Organization.

Workload types

Wherobots has the following workload types, each optimized for different use cases:

Notebooks

Interactive Jupyter notebooks for exploratory data analysis, prototyping, and visualization

Job Runs

Scheduled or ad-hoc batch jobs for production ETL, analytics, and Machine Learning workloads

SQL Sessions

Submit Spatial SQL queries programmatically to run ad-hoc analyses or power applications with a spatial SQL backend

MCP Server

Natural language interface for exploring spatial data in VS Code Copilot. MCP Server query executions are logged as SQL Sessions.

Data connectivity

Wherobots connects to your data wherever it lives.

Managed Storage

Built-in storage included with every organization

Amazon S3

Bring your own S3 buckets with cross-account IAM roles

Unity Catalog

Access Databricks Unity Catalog Delta tables directly

AWS Glue Catalog

Query AWS Glue Data Catalog tables in place

Supported languages & formats

Category	Supported
Languages	Spatial SQL, Python, Scala
Vector formats	Parquet, GeoParquet, GeoJSON, Shapefile, GeoPackage, CSV
Raster formats	GeoTIFF, COG, NetCDF, Zarr
Table formats	Apache Iceberg, Delta Lake (via Unity Catalog)
Tile formats	PMTiles (vector tiles)
APIs	STAC, REST

Organization editions

Wherobots offers three tiers to match your needs:

Feature	Professional	Innovation	Enterprise
WherobotsDB	Yes	Yes	Yes
GeoStats and map matching modules	Yes	Yes	Yes
RasterFlow	Yes Public Preview	Yes Public Preview	Yes Public Preview
SSO / SAML	Yes	Yes	Yes
Dedicated support channel	-	Yes	Yes

For full details, see Organization Editions and Wherobots Pricing.

Next steps

Create an Account

Get started with a free Professional Edition trial

Run a Notebook

Launch your first spatial analytics notebook

Explore Tutorials

Hands-on guides for every Wherobots capability

​Core products

WherobotsDB

RasterFlow

​WherobotsDB in depth

​WherobotsDB modules

​Table storage

​RasterFlow in depth

​How the products work together

​Key security properties

​Wherobots Cloud

​Workload types

Notebooks

Job Runs

SQL Sessions

MCP Server

​Data connectivity

Managed Storage

Amazon S3

Unity Catalog

AWS Glue Catalog

​Supported languages & formats

​Organization editions

​Next steps

Create an Account

Run a Notebook

Explore Tutorials

Core products

WherobotsDB in depth

WherobotsDB modules

Table storage

RasterFlow in depth

How the products work together

Key security properties

Wherobots Cloud

Workload types

Data connectivity

Supported languages & formats

Organization editions

Next steps