01

Product

Five surfaces, one substrate

A node is a complete platform instance scoped to one agency. It ships all five surfaces below. A hub runs the same five on its own data, plus federation features that aggregate across nodes. Same building blocks, two scales.

AdaptersGTFS · GBFS · TMDD · NTCIP · WZDxNormalizationschema · units · timeWarehousesystem of recordAPIs · FeedsREST · MCP · GTFS-RT · GBFSVisualizationdashboards · embeds · AI
02

What each one does

01

Adapters

Pull from the systems the agency already runs. Transit: GTFS schedules, GTFS-Realtime positions, trip updates and alerts, GTFS-Flex, GOFS, TODS, and GBFS. Roadway: TMDD 3.03d center-to-center, NTCIP 1203 signs, WZDx work zones, Waze, and 511 incidents and speeds. Weather, air quality, fleet telematics, GeoJSON boundaries, and the SQL warehouses the agency already has join the same picture.

02

Normalization

Harmonize what arrives into typed columns against the agency's own schema conventions. Units, time zones, identifiers, and coordinates resolve into one shape, so a dispatch database, a GTFS archive, and a boundary layer join in one query.

03

Warehouse

Local storage on infrastructure the agency controls: PostgreSQL as the system of record, ClickHouse for history. Any source can opt into historical capture, so every scheduled result lands in the warehouse and a record of feed history accumulates on its own.

04

APIs and feeds

REST with personal and per-query API keys, and a native Model Context Protocol endpoint for AI clients, both scoped to the caller's own permissions. Saved queries also publish as standard GTFS-Realtime and GBFS feeds, validated before every publish. Scheduled publishing, private feeds on issued tokens, and WebSocket streaming are Enterprise.

05

Visualization

Dashboards on a drag-and-drop grid, 15 chart types with a live-preview editor, embeddable widgets and public links, and Create with AI, which drafts queries and dashboards grounded in what the instance actually holds.

03

What every node ships

All of this is in the free community node, under the AGPL. The Enterprise edition adds the management layer on top: KPIs, governed reports, the alerts surface, wall and presentation modes, shared-link governance, enterprise SSO, and the AI digest.

Feed health

Every capture judged Fresh, Stale, or Down against a cadence the agency declares, with the datasets it populates named beside it.

Data catalog

Every dataset with schema, coverage, row count, freshness, and the feeds that write into it, organized into domains the instance defines.

SQL editor and visual builder

A full SQL editor with schema browser, parameters, schedules, and forking, plus a no-code builder that composes SQL from field picks. Permissions are set per query.

Spatial joins

Point-in-polygon across query results, so stops, vehicles, and boardings land in the agency's own districts, and a choropleth draws from a geometry column.

Search

One search over queries, dashboards, and datasets, with a tag vocabulary shared across object types.

White-label and deployment

Brand, palette, fonts, and domains come from configuration. Docker Compose for one machine, Helm for Kubernetes, on infrastructure the agency controls.

04

The same surfaces, two scales

An agency without a hub deploys just a node and gets the full stack against its own data, free. A region or state runs a hub: the same five surfaces on aggregated data, plus a federation layer that registers nodes, aggregates across them, and selectively pushes data back down.

Book a call