An online school that teaches the mental models behind scuba diving — not the certification. Certification only ever happens in the water, with a licensed instructor. This is the thinking that makes that training easier and safer to absorb, before and alongside it.
Why I'm building this
I've been diving for about thirty years and I'm a PADI instructor (OWSI). I learned to teach at Jernlungerne in Denmark, and I've spent liveaboard seasons in Tubbataha, based in Palawan, inside the Coral Triangle. Most of what exists online about diving is agency slide decks and dry eLearning: click through the facts, pass the quiz. The facts are there. The mental models mostly aren't.
People come out of certification able to recite the rules and are still surprised by what pressure actually does to their body, or why a reaction that keeps you safe on land can get you hurt underwater. I want to close that gap before someone gets in the water for real training — not instead of it, alongside it.
What it teaches — and what it explicitly doesn't
This teaches theory: the physics, the mental models, why the underwater world behaves the way it does. It does not teach or certify the in-water skills — breathing on a regulator, equalizing, buoyancy control, mask clearing, controlled ascent, emergency response. Those only get learned properly in person, from a licensed instructor. Nothing here certifies anyone.
Watching a video does not make you a diver. This builds mental models — it does not certify you. Scuba practised without qualified in-water supervision can cause serious injury or death.
How it's being made
I'm not a video producer or an animator — I'm a diver and an instructor building this with AI tools instead of a production budget. The curriculum comes first: I mapped the full scuba-theory syllabus into 462 individual teaching units, then classified each one by what it actually needs to be taught well — can a camera photograph the thing it's teaching, or does it need a diagram, or something drawn by hand. That classification changed the plan. Roughly a third of the material — internal anatomy, gas tables, dive-computer calculations — can never be a video, no matter how much time goes into it. It needs an interactive page instead.
Production runs through AI image and video generation, and every clip gets checked before it's used — for physics that doesn't hold together, a diver's face or gear drifting between shots, a gauge showing a number nobody verified. When something fails, the rule it teaches gets written into how the next batch is generated, so the same mistake doesn't ship twice. Right now that loop — generate, look at it properly, fix the rule, try again — is most of the actual work.
What exists today
Three films exist, and I want to be precise about their state: all three are rough cuts. None is finished, and none is published.
Video 1, "Why Scuba Diving?" (3:00) — the hook: what it feels like to enter a completely different world, and why land instincts can work against you there. Video 2, "Your Five Senses Underwater" (1:51.9) — how sight, sound, touch, and your sense of your own body change underwater. Video 3, "The Invisible Force: Pressure Underwater" (1:32) — the physics of pressure that everything else in dive training builds on.
The quality across them is uneven, and I'd rather say that plainly than not. Video 2 holds together best — consistent gear, consistent light, it reads as one dive. Video 1's individual frames are strong, but the diver's face and kit drift between shots. Video 3's generated footage is the weakest of the three, while its hand-drawn physics panels are the strongest images produced anywhere in the project so far — the tool works best exactly where I constrain it the most.
Separately, eight interactive microlearning pages are built and working — each one answers a single felt question ("why do my ears hurt?", "why can't I hold my breath and surface?") with a small on-page simulation and a short narrated clip. They're shipped internally, not public.
Of the 462 mapped teaching units, zero have a finished, shipped video. Nothing here is for sale. There's no course to enrol in, no waitlist, no launch date set.
What's next
Get one film to a bar I'm actually willing to publish, and put it in front of real people to see whether it does what it's supposed to — make someone feel like they finally get why training matters. Keep building interactive pages for the material that was never going to be a video in the first place. The public site — the place all of this eventually lives, with the curriculum as the map — comes after that, not before.