Private / Applications·Canada
North American rail

Operationalsoftware/for railroads.

// What we do

We embed with railroad operations teams, ship working software in weeks, and prepare the railroad for AI.

First working build3–6 weeks
IntegrationYour stack, untouched
Operating regionNorth America
Live · operator viewAll feeds greenUpdated 11:49 a.m.2 items need reviewPipeline nominalAuto-refresh on
The problem

Operational data is fragmented across a dozen systems that never speak to each other.

What we do

We build the layer in between — so the people running the network see what matters, when it matters.

How it integrates

Your existing systems stay exactly where they are. We read, we don't rewrite.

What you're left with

Working software, documentation, and a team that knows how to keep it running without us.

15+years
on Class I / short-line data
3–6wk
to production
7sys
shipped — asset mgmt · geometry · GIS · maintenance
1
no account managers
// Thesis

Railroads are drowning in data and starved for decisions. We build the layer in between — so the people running the network see what matters, when it matters.

Capabilities.

// Capabilities
01 / Operational dashboards

The page your team checks
before coffee.

Role-scoped views built from the data your operators already trust. Morning reports that actually get read.

ExamplesMorning report · Slow-order board · Switch audit
UsersGM · Chief Engineer · Dispatch
Ships in3–6 weeks
02 / Field applications

Tools built for the track,
the crew cab, the yard.

Hi-vis interfaces that survive gloved hands, poor signal, and cold mornings. Installed once, used daily.

ExamplesInspection app · Work orders · Compliance
UsersTrack supervisors · MoW foremen · crews
Ships in4–8 weeks
03 / AI-native workflows

Intelligence on top
of your own data.

Weather-aware planning, anomaly detection, and agentic Q&A grounded in your operational reality — not a generic model.

ExamplesGeometry anomaly · slow-order impact · agent Q&A
UsersOps leadership · planners · analysts
Ships in6–12 weeks

In use on the railroad.

// Working software
// client · rail ops / morning
Morning operations dashboard
Rail · Ops

Morning Report

Saturday, April 11 · 06:00 — 12:00

Slow orders · region view →
1242
MainlineBranchYard
Line 03 — West · 2.1 mi active
Open defects view →
7138
PriorityAgingNew
Inspections · 24h view →
100
ExceptionReviewedPending
32 reports · updated 11:49 a.m.
Morning report · desktop + mobile
GeoHUD · track geometryGeoHUD track geometry viewer
Year-over-year gauge, cross-level, curvature · Geometry review
God's Eye · network viewGod's Eye network map
Defects, slow orders, speed restrictions — one map · Network map
// Approach

How we work.

Three principles, in order. Skip the first and the other two stop working.

01 / Embed · 02 / Ship · 03 / Leave behind
01
[Discovery]

Embed with the people running the railroad.

We sit with operators, ride the crew cab, and watch the morning report go out. Two weeks. No slides.

Duration1–2 weeks
WhoOps · Field · Engineering
You getWritten findings + build plan
02
[Delivery]

Ship something working in weeks, not quarters.

First working build in 3–6 weeks. Iterated with the people who use it every shift. No staging lanes, no change-management theatre.

Duration3–6 weeks to prod
Integrates withYour stack, untouched
You getLive software + docs
03
[Handoff]

Leave behind a team that can keep it running.

We document, pair with your engineers, and hand off clean. No vendor lock-in. No account managers chasing renewals.

Duration2 weeks handoff
WhoYour internal team
You getRunbooks + source + zero retainer
// Engage

If you run a railroad,
let's run it sharper.

Short briefing. No pitch deck. We show you where we'd start, what we'd ship in six weeks, and what it would cost to find out we were wrong.