Meteorological data · Software · Analytics
We are experts in meteorological data and committed to using technology to improve the wisdom of society.
We build meteorological software, apps and analytics for companies — and we have built apps for major global companies.
Built for
major global companies
What we build
Four things, done properly: the software, the apps, the analysis behind the numbers, and the plumbing that keeps all of it fed with good data.
Forecast and decision-support tools built around your operation, not around a generic weather widget. Parameters, thresholds and views that match how your team actually works.
Consumer and professional apps that make meteorological data readable in the field — offline-capable, fast, and clear enough that nobody needs a meteorology degree to use them.
Statistics, model verification and bias correction that answer the only question that matters: how much can you trust the next number, and what is it worth to you?
Ingest, quality-control and serve meteorological data from many sources through one dependable interface, so your systems always read from the same version of the sky.
Where the data comes from
Weather data arrives from models, instruments and people — each with its own bias, resolution and failure mode. Our job starts with handling that honestly.
We pull from models, radar, satellites and ground networks, and keep every source on its own clock and its own failure mode.
Range checks, spatial consistency, climatology and cross-source comparison. Bad data is caught before it reaches a decision.
Statistical correction and downscaling where it helps, then continuous verification against what actually happened.
The result ships as an app, a dashboard or an API — with the uncertainty visible instead of hidden behind a single confident number.
Illustrative interface
Apps in the field
A forecast is only useful when someone can read it in ten seconds, on a small screen, in bad light, often without signal. That is the bar we design against — then we make the data behind it hold up.
Production forecasts and market views where a few percent of accuracy is a real number at the bottom of the page.
Route and schedule risk: where the weather will close a road, delay a port or ground a flight, and how early you can know.
Publishable forecasts at scale — the same underlying data, dressed for a screen and a viewer who gives it three seconds.
Crews, events and maintenance planning, where the answer has to work on a phone with two bars of signal.
How we work
We start from the meteorology and the business question, then choose the technology. It keeps us from shipping clever systems that answer the wrong thing.
Every forecast we deliver can be compared to what actually happened. Verification is part of the product, not a later report.
Data sources change, formats break, models get updated. We build for the maintenance phase, because that is where weather software spends its life.
About
Metcube is a specialist meteorology shop — small, senior, and organised around one skill: turning weather data into products that hold up when someone actually depends on them.
Most weather software is built by people who have never had to defend a forecast. We work the other way round: the meteorology comes first, and the software is built to serve it. That changes what gets prioritised — how uncertainty is shown, how a bad observation is caught, what a forecast is allowed to promise.
We take on the whole span of that job: the ingest and quality control behind the scenes, the analysis that decides what a number is worth, and the app or dashboard a real person reads with ten seconds of attention. Keeping all three in one team is what makes the result coherent.
CEO
Entrepreneur and meteorologist
Andreas leads Metcube, combining a meteorologist's relationship with the data and an entrepreneur's focus on shipping something companies will actually use. He has taken meteorological products from a blank page to apps in the hands of users at major global companies.
Contact
Send us the decision you are trying to make — a production forecast, a route risk, an app for your crews — and we will tell you honestly whether data can help, and how.
