Artificial Geospatial
General Intelligence
Multi-agent intelligence for the physical world
AG2I is multi-agent geospatial intelligence — a geospatial codebase built from the ground up, over a million lines of code, purpose-built for Anthropic's Claude Code, to process anything geospatial as a coordinated, self-verifying team of agents. Fully automated, adaptive command-line and GUI workflows designed around your data, not the other way around. What is geospatial general intelligence? →
The first AI-native geospatial processing system.
Conventional geospatial software forces every dataset to conform to a fixed, rigid workflow. AG2I inverts that relationship. By embedding Anthropic's Claude Code directly into a purpose-built, million-line geospatial codebase, the system reasons about each dataset and composes the pipeline to fit it — automating the analyst's judgment rather than merely executing fixed steps. The result is a general capacity to process anything geospatial: aerial, oblique, LiDAR, satellite, terrestrial, or street-level.
Claude Code at the core
An integrated reasoning agent interprets the data, selects methods, and orchestrates the toolchain — turning an expert workflow into an automated one.
The system adapts to the data
Both the command-line engine and the GUI reconfigure themselves around each dataset's geometry and sensor model — not the other way around.
Built from the ground up
More than a million lines of original, modular code — engineered to expose the full power of the most capable geospatially trained AI available.
Survey-grade accuracy, fully automated.
Beneath the intelligence layer sits rigorous geodetic science: aerotriangulation, trajectory compensation, and bundle adjustment at scale — the core of aerial sensor processing. The system models time-sync offsets, lever-arm geometry, and platform motion to resolve raw imagery and point clouds to true, reportable accuracy — and carries that accuracy from the airborne sensor all the way to the ground.
Aerotriangulation
The foundation of aerial sensor processing: high-redundancy, multi-ray tie-point networks solved by robust bundle adjustment, yielding sub-pixel image residuals across thousands of exposures.
Trajectory Compensation
Refinement of platform trajectories through time-synchronization and lever-arm correction — the basis for oblique triangulation and air-to-ground accuracy transfer.
Adaptive Automation
A fully automated, adaptive CLI — with a GUI designed around the dataset — that configures and executes each processing stage without manual intervention.
A unified, modular processing stack.
Each engine is an independent module with a rigorous methodological basis; together they compose a continuous pipeline from raw acquisition to interactive, machine-readable 3D models.
OpenGlobe
A WebGL virtual globe rendering orthoimagery, terrain, vector HD maps, and point clouds within a geodetically rigorous reference frame.
OpenGeoserve
Server-side GPU rendering and pixel streaming that decouples computational load from client hardware for thin web and native viewers.
OpenLiDAR
Full raw-to-product LiDAR processing — from raw airborne and terrestrial returns through calibration, strip adjustment, and classification to finished deliverables — with hierarchical octree indexing for view-dependent streaming of billion-point datasets.
OpenStereo
Dense multi-view stereo photogrammetry deriving digital surface and terrain models through epipolar resampling and sub-pixel image correlation — the foundational core.
OpenStreetX
Full raw-to-product processing of street-level imagery and mobile LiDAR — calibration, georeferencing, and registration of ground capture to the aerial block — producing finished street-level datasets that feed automated feature extraction.
OpenFeatureX
Fully automated geospatial feature extraction from high-accuracy datasets, purpose-built for GPU architecture.
OpenLineWorks
Geometric registration of vector and raster data via thin-plate-spline warping and iterative correspondence refinement under non-rigid deformation.
OpenVoxelite
Volumetric fusion of orthoimagery, LiDAR, and HD Map data into a discretized voxel field supporting spatial query and occlusion reasoning.
OpenNavHD
High-definition mapping and positioning framework providing the lane-level geometric substrate for navigation and dead-reckoning.
OpenGeoLook
Interactive inspection and conditioning of heterogeneous datasets, preparing raw inputs for automated processing and AI-driven reasoning.
OpenGIS
A geospatial workbench for inspecting, styling, and analyzing vector and raster data — the interactive surface across the suite, built on open foundations.
OpenMultiAgent
An orchestrator–worker multi-agent system that plans, dispatches, and verifies large-scale geospatial production — a synthetic workforce that runs the whole pipeline.
Certified HD maps for autonomous vehicles.
HD mapping for autonomous vehicles is a demonstration of what a general geospatial processing system can produce — not the limit of it. The same automated pipeline generates certified, continuously verified map tiles, fused with dead-reckoning, that give a vehicle a trustworthy and measurable model of the road.
Because the underlying capability is general, the system applies equally to mapping, surveying, infrastructure, environmental monitoring, and any domain where the physical world must be measured accurately.
- True absolute accuracy with reports — measured and certified, not estimated.
- Dead-reckoning continuity through GNSS dropouts and tunnels.
- Continuous, automated QA keeping the map fresh and verifiable.
- Tile-based & streamable — only the world you need, when you need it.
- High-fidelity NavHD datasets — the highest form of lane-level navigation.