Build Build with the AI tools you already know, now spatially enabled. No new languages. No new tools. Write SQL and Python in you preferred AI enabled IDE: Cursor, VS Code, Windsurf, Claude Code, or through the Spatial AI Coding Assistant. WherobotsDB or RasterFlow enables the physical world data processing. START BUILDING Spatial AI coding tools Discover datasets, construct queries, and run production jobs through natural language. From notebook to production in one platform Explore in notebooks. Deploy with job runs and Airflow. SQL and Python, nothing else to learn Standard spatial SQL and Python APIs. The Spatial AI Coding Tools: Your AI copilot for physical-world data The Spatial AI Coding Tools connects AI agents to spatial data. Through VS Code, Wherobots Command Line Interface (CLI) and Model Context Protocol (MCP), agents can discover spatial datasets, construct spatial queries or notebooks, and execute production workflows in natural language. Available on the VS Code marketplace and compatible with Claude Code, Cursor, Windsurf, OpenCode, and other MCP-enabled coding tools. TRY IN VS CODE Build spatial AI from notebooks to production Four interfaces into the same spatial compute engine. Explore in notebooks, query with the SQL API, integrate with the Typescript SDK, or automate with Job Runs. Notebooks SQL API Typescript SDK Job Runs Serverless Compute and AI Join vector, raster, and structured data in one platform WherobotsDB processes vector geometries, satellite imagery, and structured tables in a single query. No separate tools for raster and vector. No export steps between systems. Spatial joins, ETL, and transformations run at planetary scale. Raster+Vector ETL & Joins Planetary scale TALK TO US 300+ spatial functions for vector and raster data 300+ spatial functions covering vector and raster data, with native Spark SQL for tabular operations. Advanced spatial analytics include DBSCAN clustering, Getis-Ord Gi* for hot spot detection, and Local Outlier Factor for anomaly identification. All accessible through standard SQL and Python. DBSCAN Getis-Ord Gi* LOF 300+ fns LEARN MORE Start from real-world solutions Solution notebooks for insurance risk scoring, agricultural field detection, mobility analytics, and infrastructure mapping. Each notebook contains working code on real data. Start from a proven example and adapt it to your use case. BROWSE NOTEBOOKS Write spatial AI in the tools you already use Python, SQL, Scala, and Java in Jupyter notebooks. The Spatial AI Coding Tools in VS Code. MCP and CLI integration for Claude Code, OpenCode, or elsewhere. The same spatial functions, the same WherobotsDB or RasterFlow engine, in any interface. Python SQL VS Code LEARN MORE 100% Apache Sedona compatible Existing Apache Spark and Sedona workloads run on Wherobots Cloud 3X faster with zero code changes. GeoPandas workflows scale to planetary scale datasets through the Sedona API. Built by the original creators of Apache Sedona. Spark API Sedona API GeoPandas API LEARN MORE From notebook to production in one step The same code you write in notebooks run as automated jobs in Apache Airflow or other orchestration tools. No rewrite needed. No separate deployment pipeline. Airflow Scheduling LEARN MORE Dive Deeper AI Agents for Spatial Analytics: A 15-Minute Experiment with Our MCP Server A hands-on look at using the Wherobots MCP server to go from a natural language question to real spatial query results in minutes. READ MORE Scaling Spatial Analysis: How KNN Solves the Spatial Density Problem for Large-Scale Proximity Analysis A walkthrough of processing 44 million geometries across 5 US states using kNN joins to solve the spatial density problem that breaks traditional proximity analysis. READ MORE The Medallion Architecture for Geospatial Data: Why Spatial Intelligence Demands a Different Approach How to apply the bronze/silver/gold pipeline pattern to spatial data, and why geospatial workloads demand a different approach than standard analytics pipelines. READ MORE FREQUENTLY ASKED QUESTIONS What is the Spatial AI Coding Assistant? The Spatial AI Coding Assistant connects AI agents to spatial data through VS Code and Model Context Protocol (MCP). Agents discover spatial datasets, construct spatial queries, and execute production spatial AI workflows in natural language. It is available on the VS Code marketplace and compatible with Claude Code, OpenCode, and other MCP-enabled coding tools. How does Wherobots accelerate my spatial data workflows? WherobotsDB is a high-performance, cloud-native engine optimized for spatial data (vector and raster). The engine processes planetary-scale datasets up to 20x faster and at a fraction of the cost of general-purpose cloud engines. WherobotsDB uses standard SQL and Python, so data teams build spatial AI applications without learning new tools or languages. What languages and developer frameworks are supported in Wherobots? Wherobots supports the most common data and spatial development interfaces. You can develop using: Spatial SQL Python Scala Java The platform offers a seamless development experience through familiar Jupyter notebooks and integrates with existing data workflows through REST APIs, client SDKs (e.g., Python driver, Java JDBC driver), and an Apache Airflow provider for pipeline orchestration. Get started WherobotsDB is part of Wherobots, the AI Context Engine for the Physical World. TALK TO US START BUILDING
AI Agents for Spatial Analytics: A 15-Minute Experiment with Our MCP Server A hands-on look at using the Wherobots MCP server to go from a natural language question to real spatial query results in minutes. READ MORE
Scaling Spatial Analysis: How KNN Solves the Spatial Density Problem for Large-Scale Proximity Analysis A walkthrough of processing 44 million geometries across 5 US states using kNN joins to solve the spatial density problem that breaks traditional proximity analysis. READ MORE
The Medallion Architecture for Geospatial Data: Why Spatial Intelligence Demands a Different Approach How to apply the bronze/silver/gold pipeline pattern to spatial data, and why geospatial workloads demand a different approach than standard analytics pipelines. READ MORE
