BeyondView

A rendering problem wearing a real estate costume

Here is a fact about commercial real estate that nobody puts in the brochure: an empty office floor is very hard to sell, largely because it is empty. It has exposed ceiling grids, a lonely fire extinguisher, and the specific fluorescent sadness of a space nobody has moved into yet. Landlords know this, which is why they spend real money on staging, demolition, and getting prospective tenants to physically walk the floor and imagine something better.

BeyondView, a small company in San Francisco, looked at all of that and decided it was, at heart, a rendering problem. If the issue is that a tenant cannot picture the space, you do not necessarily need to build the space. You need to build a convincing picture - an accurate, interactive, photorealistic one - and let the tenant walk through it from a laptop. The company calls these "virtual spec-suites." The rest of us would call them digital twins: software copies of physical buildings, accurate enough to measure and pretty enough to lease.

The mechanically interesting part is how BeyondView makes them. A lot of the industry gets a digital twin by sending someone to the building with a laser scanner, which is fine but requires the building to exist and someone to go to it. BeyondView's pitch is that it can build the twin from prepared materials - blueprints, photos - using a stack of computer vision, machine learning and deep learning running on a cloud platform. No scan truck. In principle, that is cheaper and faster, which in commercial real estate is roughly the entire ballgame.

That one line does a lot of work. "Market" is the obvious part: prettier listings, 360-degree panoramas, video walkthroughs, 3D maps. "Manage" is the more ambitious part, and it is where BeyondView is really trying to live - a portfolio management portal where owners and operators can view, modify and monitor properties across devices, with engagement metrics attached to every virtual tour. If you are a landlord, this is the difference between a marketing gimmick and an operating system.

A Wikipedia guy, a Pixar guy, and the man who greenlit Iron Man

The founder is Kul Wadhwa, who is not the person you would predict for a commercial real estate startup. Before BeyondView he spent nearly seven years in business development at the Wikimedia Foundation - the nonprofit behind Wikipedia. There is a certain logic to going from "make the world's knowledge free and open" to "make the world's buildings visible and open," even if the balance sheets look nothing alike.

The rest of the roster is where it gets genuinely fun. BeyondView's team pulls from Stanford's HSTAR Lab, Electronic Arts, Marvel Entertainment, Disney/Pixar, Linden Lab, Samsung, Wikipedia and SAP. According to press coverage, a former Pixar build engineer, Jeremy Yabrow, developed early product technology; George Borshukov, a former director of creative R&D at Electronic Arts, contributed; and Peter Cuneo, the former Marvel CEO who greenlit the Iron Man franchise, has been associated with the company as chairman. Commercial Observer ran the whole thing under the headline about people who helped bring 'Avatar' to life now making digital twins, which is the kind of headline you cannot buy.

There is a real thesis buried in that gag. Visual-effects talent is rare, expensive, and trained to make imaginary spaces look photographically real under deadline. Movie work is lumpy. Point that same skill set at an industry - commercial real estate - that has always struggled to show its product, and you have a small team that can do something most proptech companies cannot: make the fake floor look like the real floor. Twelve people, roughly. It does not need to be a hundred.

Why "cheaper" is the whole strategy

The competitive landscape here is not empty. Matterport, Cupix, Giraffe360 and a handful of others all sell some version of "here is your building, but digital," mostly built on physical capture - you scan the space, you get the twin. BeyondView's differentiator is the claim that it can skip the scan and generate the twin from blueprints and photos with AI, at lower cost. If that claim holds at scale, it is a meaningful edge, because in real estate the party that shows the space more cheaply tends to win the party that shows it more beautifully.

The WeWork Japan deal is the tell. WeWork's entire model is flexible space that has to be marketed and re-marketed constantly, across a large portfolio, ideally without flying a photographer to every location. That is precisely the use case where a scan-free, cloud-based digital twin factory earns its keep. It is also the kind of enterprise agreement that a twelve-person company signs when its technology is doing something the customer genuinely cannot do in-house.

So the honest way to describe BeyondView is not "the company that will disrupt real estate." It is smaller and more specific than that, which is a compliment. It is a company that took a rare, cinematic skill - making imaginary rooms look real - and aimed it at an industry that has always paid handsomely for exactly that, without knowing it could ask a rendering company for it. Whether that becomes a large business or a very good acquisition, the underlying observation is sound: a lot of real estate's hardest problems are, on inspection, pictures that haven't been drawn yet.