VIA — construction field verification infrastructure
Reality Capture · Part One

The Scanner That Has Nowhere to Go

What started as a three-sided marketplace six months ago has been rebuilt more times than I can count. Last week, after all of it, the work finally touched a real building.

Michael Sturgeon · Founder, VIA Reality Capture Series

What started as a three-sided marketplace six months ago has grown, transformed, and gone through about as many iterations as I have had sleepless nights. Anyone who has built something from nothing knows the particular kind of restless that comes with it. You build, you break it, you test, you break it again, and the whole time the industry you are building for keeps advancing faster than your own builds can keep up. That gap, between where the work is and where you need it to be, will drive a person wild.

And at some point you start to wonder whether the thing on your screen holds up the second it touches a real building.

Last week I found out. We ran our first field scan. After months of theory and rebuilds, the work became real, and getting back out with a crew, in the industry I have been part of for twenty years, was the most grounding thing I have felt since I started.

So before I say anything about the technology, I want to talk about what I saw, because it is the whole problem in one picture.

01 / The patternEveryone is capturing. Nobody is using.

Reality capture is everywhere in construction right now. Scanners, drones, 360 cameras, phones. The industry is generating an extraordinary amount of data about what physically exists on a jobsite. Walk almost any large project and someone is capturing something.

And then almost none of it gets used.

The scan gets delivered as a file. A beautiful, accurate, dense file that sits on a drive. It does not tell anyone whether the work matches the drawings. It does not confirm scope. It does not back a pay application. It does not settle the argument that shows up in every closeout meeting about what was actually built versus what was designed. It is a perfect photograph of a jobsite that answers no question anyone is being paid to answer.

The capture problem is mostly solved. The "so what" problem is wide open.

That gap, between what the drawings say and what exists in the field, is not a stone problem or a concrete problem or a steel problem. It runs through the entire construction ecosystem. The as-built versus as-designed gap is where disputes live, where money gets stuck, where trust between an owner, a GC, and a trade quietly erodes. And right now the record that should close it is either photos buried on someone's phone or a point cloud nobody can read.

VIA exists to close that gap. To take trustworthy field capture and turn it into a structured, auditable record that every stakeholder on a project can actually use. Not a photo. Not a buried field report. A record you can stand behind.

That is the technical story, and I will get into the hardware side of it in the next piece. But none of it happens without people willing to bet on an idea before it is proven. So this first one is about them.

02 / The peopleNobody builds this alone

An idea on a screen is just an idea until someone gives it a place to stand. A few people did exactly that, and I am not going to pretend the credit is mine alone.

With thanks

Jeff Brekhus and Scott Polak at Brekhus Tile & Stone — for opening the door and letting VIA run a pilot on real work. I spent years inside that company's operations, and the trust to come back with something new and unproven is not a small thing. It is the kind of bet that makes a company like this one possible.

Jason Dysthe at Valhalla Instruments — for being the right hardware partner at the right time. Jason brought the reality capture front end to the table and has been a genuine collaborator, not just a vendor. The supply chain behind a construction technology decision matters, and this one was built on it.

What struck me most last week was how much of this is teamwork. The trade that knows the work. The hardware partner who knows the capture. And the layer I am building that has to make the two add up to something a project can use. None of those three parts means much alone. Together they start to look like the way construction should have been able to verify itself all along.

Twenty years in, I am more convinced than ever that this industry deserves better tools than the ones it has been handed. I am glad to be back out where the work actually happens, building one.