It’s hard to teach industrial maintenance when there’s only one machine for thirty students. We put the machine in their pocket.
That’s a clean line. The actual problem took longer to articulate. But that’s the core of it: vocational education is inherently hands-on, and hands-on doesn’t scale the same way a lecture does. You can put thirty students in a room and show them slides about a CNC machine. You can also put one CNC machine in a room and have thirty students take turns for a few minutes each. Neither of those is teaching industrial maintenance.
SENAI runs vocational programs across Brazil — welding, electronics, industrial automation, building systems, health sciences. The schools vary enormously in resources. A school in São Paulo might have full workshop facilities. A school in a smaller city might have one of each piece of major equipment, shared across multiple class sections, multiple years. The equipment gap between schools is real, and it shapes what students can actually learn.
What AR Actually Changes
The obvious pitch for AR in education is “it’s cool.” That’s not the interesting part.
The interesting part is what becomes possible when every student has their own instance of the thing being taught. Not a shared machine with a queue. Not a diagram. Their own manipulable, zoomable, activatable model of a centrifugal pump or an electrical switchboard or an automotive engine — overlaid on the real-world surface in front of them, at full scale, right through their phone camera.
SENAI Space lets students do things you can’t do with a physical machine in a classroom. You can activate an animation showing internal fluid flow. You can peel back layers to see components that are normally hidden inside casings. You can freeze a process mid-cycle and rotate the model to look at it from underneath. None of that is possible with the real thing without expensive teardown or simulation software that requires dedicated lab computers.
How It Worked in Practice
The intended classroom flow was deliberate: this wasn’t meant to replace instruction, but to slot into it.
A teacher introduces a concept — say, the internal combustion cycle. Students open SENAI Space, point their phones at a flat surface, and a full-scale engine appears. They can rotate it, zoom into the piston assembly, trigger the animation showing the four-stroke cycle. Then the discussion happens differently than it would with a diagram, because everyone has been looking at the same thing from their own angle.
The app covers 21 knowledge areas — engineering, electronics, industrial maintenance, building systems, health, agribusiness, and more. The 3D model library spans equipment categories that between them represent most of what SENAI teaches.
🚧 Need more context: Which equipment categories were most used? Was there data on which 3D models got the most interactions — and did that correlate with which knowledge areas had the worst equipment access in physical schools?
The Equity Argument
I want to be honest about what AR does and doesn’t solve for educational equity.
It doesn’t replace a physical internship. It doesn’t give students practice on real controls with real consequences. A student who has only ever used the AR model of a welding machine is not the same as a student who has spent forty hours at a welding station.
What it does is change what “show, don’t tell” costs. Before, showing a student the internal components of a hydraulic press required either having the press and doing a teardown, or having very good static diagrams. After, showing that same student the same components required a phone and an open app. A school in a smaller city with limited equipment suddenly had show-don’t-tell capability in 21 knowledge areas.
The app launched free on iOS and Android, which mattered. No per-school licensing, no lab software installation, no dedicated hardware. Students could use it at home.
🚧 Need more context: Were there student usage numbers? Did teachers report using it outside of formal class time — for student self-study? Any classroom adoption data or teacher feedback from the rollout?
What I Took Away
Building SENAI Space taught me something about the difference between features and enabling conditions. The AR technology was the feature. The enabling condition was that it ran on phones students already had, in classrooms that didn’t need to be reconfigured, in subjects where the alternative was genuinely inadequate.
The best product decisions I made on this project were the ones that kept the barrier to entry as low as possible. No account required to view. Available on both major platforms. Free. If we’d required a SENAI login or a dedicated device, we’d have built something impressive that nobody used.
🚧 Need more context: What was the development timeline? How many 3D models were in the library at launch vs. now? Were models created in-house or with external 3D artists?