We’re Human After All

My last two posts were about the machine: the platform agent-native BIM requires, and where it sits in the world of tools and data you already run on. Today is about the people, because that's who all of it is for. Strip away the architecture and the acronyms, and what we actually built is a place for people to do their best work. All people, not just the ones with a decade of software scar tissue. Here's what that means at the keyboard.

Software this important should feel light

Ask designers what they want from their tools and you'll hear the same words again and again. Light. Simple. Fast. It should get out of the way. It should keep you in flow. It should feel like an extension of how you think. Nobody has ever described BIM software that way, and everybody knows it.

For a generation, the power of these tools has belonged to the specialists who spent years learning to fight them. Everyone else waited outside: the principal with the sharpest eye in the firm, reduced to marking up printouts. The client who has to live with the result, squinting at a screenshot. The junior designer with the idea nobody asked for, told to learn the software first. The tools became a gate, and the gate decided who got to participate in design.

We built Motif to take the gate down. It opens in a browser: no installs, no IT tickets, no six months of training before you're allowed to touch the work. The interface everyone already knows is their own language: describe what you want, in plain words, and the model responds. Want the corridor wider? Say so. Want to see the lobby in the afternoon light? Ask. If you can say it, you can shape it. That's not a convenience feature. It's the difference between software for operators and software for everyone with something to build.

Simplicity like that is the hardest thing to engineer. Making complexity available is easy; every legacy tool proves it. Making complexity feel obvious, without hiding the capability, took rebuilding everything underneath. It's also the most important thing we do, because simplicity is what decides who actually gets to participate.

The gruntwork goes away

The other half of feeling light is what you no longer carry. The schedules, the repetitive sheets, the checking against requirements, the chasing of answers: agents handle the toil in the background while you design. Your firm's standards and hard-won know-how become skills the agents apply automatically, so the work comes out your way without anyone policing it. And the judgment never leaves your hands. Agents don't decide what a space wants to be or when a plan is done. You do. More of you, not instead of you.

At speed and scale

Speed is the first thing you'll notice, and it comes from how the platform is built. The full model runs on powerful infrastructure while the browser becomes a fast window into it, keeping the experience fluid whether you're placing a doorknob or reworking a floor plate. As the design changes, plans, sections, models, and sheets update in real time. No progress bars to watch, no sync to wait on, no splitting the building into fragments to keep the software happy.

Scale is the quieter half of the promise. A real project is enormous, and Motif holds all of it: the whole model, every relationship, live for every person at once. No worksets to manage, no linked files to reconcile, no waiting your turn.

And speed extends to what you see. We point the same GPUs that render the worlds of modern video games at your building: real-time, fluid, genuinely beautiful, not a laggy wireframe from another era. Architects design gorgeous buildings on software that looks like it wandered out of 1998, and that's always struck me as backwards. Your buildings are beautiful. Your software should be too.

It just has to work

Simple, fast, beautiful. Underneath all of it sits one standard, and it's the one this industry has been denied for too long: it has to just work.

That means doing the real work. There's a vast gap between a slick render that wows a room and a coordinated, code-passing set someone can actually build from, and this is where most AI in our field falls apart. The real work means a wall that behaves like a wall, with a type, a thickness, a fire rating, and documents that hold together across every discipline and survive the plan check. Real-world impact is the only test we accept, because it's the only one you can accept.

And it means being dependable. No crashes on the deadline, no ritual of workarounds, no fighting the thing that's supposed to be helping you. You should be able to trust it with the project on the worst night, when it counts. Ease means nothing if you can't rely on it. A tool for everyone has to be worthy of anyone.

The climb

None of this is easy, and I won't pretend we've finished it. We haven't. But we've started the climb. The journalist Martyn Day has a phrase for this project: going up the north face of the Eiger. It's the hard way, the route most people look at and decide isn't worth the risk. But someone has to do it, and it's a problem worth solving.

What we've shipped is a great first step. Not the summit, but a real, honest beginning. The feedback has been better than we had any right to hope for, and it's shown us exactly how much is still ahead. We're accelerating, and we're not doing the hard version to prove we can. We're doing it because the people who design, engineer, and build the world deserve tools worthy of the work, and for a long time, they haven't had them.

The best part of the climb is still in front of us.

‍