Cartos · Phase II — In development

The Analyst Intelligence Layer.

Phase I made Cartos see and stream the world. Phase II makes it reason about change, build in 3D, derive new analysis-ready data, and write its own auditable conclusions — turning a visualization substrate into one that produces intelligence.

The shift

From presentation to production.

Today an analyst studies imagery, then leaves the platform to do the thinking — change detection in one tool, 3D in another, the brief in a word processor. Cartos pulls that whole workflow inside, and hands most of it to the agent.

01

See & stream

Own & open imagery, telemetry, and 3D in one sovereign globe. today

02

Reason & derive

An AI analyst runs the analysis and produces governed, reusable data over your geography.

03

Report & decide

One click turns a session into a branded, signed deliverable — methodology written for you.

The capabilities

Five new capabilities. One intelligence loop.

IDCapabilityWhat you can showRole
M03GeoAI Agents & Derived DataAgents mint reusable, governed datasets over an AOI that compound across sessionsThe moat
M02Comprehensive ReportingA session → a signed HTML / DOCX / PDF report with an auditable methodology sectionThe deliverable
M01Temporal Change DetectionTwo epochs over an AOI → a georeferenced change layer with confidenceThe analytic
M04Gaussian Splat 3D map viewer livePhotoreal, georeferenced 3D served GPU-free inside the map — incl. SuperSplat SOGS importReality capture
M05AOI → 3D BuildingsReconstruct textured LOD 2.5/3 buildings from imagery, with an honest accuracy labelDigital twin

Reality capture · live now

Photoreal 3D, served on the map.

A 3D Gaussian splat reconstructs the real world from ordinary photos or drone imagery as millions of soft, coloured points — photoreal, not a simplified model. Cartos serves them GPU-free in the browser, anchored in their real location: fly through a site, a corridor, or a landmark, then pitch to ground level and walk it.

A photoreal 3D Gaussian splat of Red Rocks Amphitheatre (ESRI drone sample) served on the Cartos map, with the 3D Data panel open.

Capture reality cheaply

A phone or drone flyover becomes a navigable 3D asset — no LiDAR rig, no manual modelling.

On the emerging standard

Imports the compressed, streamable SOGS bundle (SuperSplat / PlayCanvas) — level-of-detail tiling, ~15–20× smaller than raw — plus .splat / .ksplat / .ply / .spz.

Sovereign & light

Client-side WebGL serving: no server GPU and nothing leaves your perimeter to view it. Generation is the only GPU step, and stays gated.

How it works

Build the loop before the spectacle.

The photoreal 3D wins the room — but the loop wins the contract. Reusable intelligence and auditable reporting are the connective tissue that make every other capability worth paying for.

ObserveOWN & OPEN IMAGERY ReasonAGENT-DRIVEN GEOAI DeriveGOVERNED DATASETS ReportSIGNED DELIVERABLE DERIVED DATA COMPOUNDS — EACH SESSION FEEDS THE NEXT
Figure 2. Observe → reason → derive → report — and the derived-data graph feeds back into the next session.
“Cartos is the sovereign geospatial intelligence platform that reasons, builds in 3D, and documents itself.”

Delivery

Sequenced in waves — value ships continuously.

Build the loop (derived data + reporting) first, then the analytic, then the 3D spectacle. Every capability ships independently behind a flag. Figures are planning estimates in engineer-weeks, not quotes.

0102030 engineer-weeks Wave 2AM03 + M02 · the loop 26 Wave 2BM01 · change detection 12 Wave 2CM04 + M05 · 3D 24
Figure 3. Build effort by wave — ~58 engineer-weeks total (planning estimate for sequencing).
WaveModulesGoalEffort
2AM03 derived data + M02 reportingRun an analysis → walk away with a finished, auditable report~26 eng-wks
2BM01 change detectionFirst high-value analytic, riding the derived-data graph~12 eng-wks
2CM04 splats → M05 3D buildingsThe visual differentiator, packaged by the report engine~24 eng-wks
TotalFive capabilities, independently shippableOne AOI, end to end — reconstruct, detect change, generate the report~58 eng-wks
Infrastructure

Async, cost-honest compute

Heavy AI and 3D work runs as asynchronous jobs on a GPU tier that reports its own cost — the platform never blocks, and spend is observable from day one.

Sovereign

In-territory by default

Model weights and compute stay inside your boundary; non-sensitive work can optionally burst, but sovereign is the default route.

Trust

Auditable by construction

Every report's methodology is generated from the work that was actually executed — confidence on every change layer, an honest grade on every reconstruction.

Design partner program

See the loop on your own AOI.

The earliest wave delivers the signed-report moment — the fastest way to prove the value on real geography. Engagements run under NDA.