← The build Svenska

The build · The goal

I'm building the world's best workout app. So far it has made 309 kronor.

I've had the goal for fifteen years. I'm saying it out loud now because of everything that has gone wrong along the way, and because I intend to show where the app stands, point by point.

I have been training for twenty-four years, and for fifteen of them I have wanted to build a workout app that is better than all the others. For almost two years I have been trying for real. The result so far is three attempts from zero, four names for the app, a spam verdict from Apple and 309 kronor and 72 öre in revenue, just under thirty US dollars. Almost half of it came in one day, when the app was given a price by mistake.

That is why I'm saying the goal out loud. Anyone who claims to be building the world's best anything sounds like a commercial, and the only thing that separates an ambition from a commercial is showing the whole road, including the embarrassing part. That part is in the timeline, with pictures.

What does best mean?

Best is not the most features. A workout app is best if it gets you to train tomorrow as well, and in ten years. Everything else is a means. I measure MassBuilder against five things.

1. Logging must not be in the way

You are standing there with chalk on your hands and your pulse in your throat. A set should be one tap: reps, weight and time are saved, and the rest timer starts on its own.

2. The app should know what you do next

Your next weight is a function of what you managed last time, not of what you feel like. Make all your reps and the weight goes up. Miss the same weight twice and the app suggests a lighter one, and after three misses it lowers the weight itself.

3. It should be where you are looking

The same set shows on the phone, the watch, the iPad and the TV at the same time. I haven't seen another workout app do that.

4. Your training is yours

No account, no ads and no tracking. Your log lives in your own iCloud, and I can't see it. That costs me something: I don't know where people get stuck in the app, because I have promised not to look.

5. It should make itself unnecessary

The thread through the app is four steps: motivation, discipline, routine and identity. Motivation is worthless to rely on. Discipline is the hardest part. Routine is where it starts to be fun. Identity is when you are a person who trains, and then you don't need the app any more. I want to build an app that can live with that.

Your lifestyle, as a tree: Build.
Build, new in version 3.0: your training as a tree, from motivation to identity.

Where it measures up

  • One tap per set, on all four screens.
  • The daily challenge: an alarm that only goes quiet when the camera has counted your reps.
  • A form check for eight lifts, where every rep is played back next to the correct movement.
  • 15 programs, 317 workouts and more than 400 exercises as 3D figures.
  • Fuel: say, type or photograph what you ate, and the app counts the macros and gives a weight forecast with an honest range.
  • Build: your training as a tree that grows with your routine and never dies.
  • The Bench: five judges with opinions about how you train, never about how you look.

Where it doesn't yet

This is the list a commercial would have skipped.

  • Almost nobody has rated it. Seven ratings in Sweden, all five stars, and one in the US. That proves nothing yet.
  • It only exists for Apple devices. No Android and no web.
  • Four languages: English, Swedish, German and Spanish.
  • Nobody has paid for Pro. Through November 10, 2026 the app had been downloaded 392 times. 241 got Pro for life for free with a code I handed out myself, and three had started a trial.
  • The camera is a kind judge. The counter in the daily challenge would rather let a doubtful rep through than fail a real one, because nobody should have to pay because the camera can't find the right angle in a cramped flat.
  • I have never sold anything. I have been building for twenty months and selling for none. That is the next thing I have to learn, and it is in the post about daring to sell.

Why it is possible now

When I started I couldn't code, couldn't design an interface, couldn't build a database and couldn't lay out an architecture. I got better with every attempt. What decided it was something else, though: AI got better even faster.

In the spring of 2025 I got an app with a watercolour background that I wrote FULT across myself, which is Swedish for ugly. In the autumn of 2026 I work with an AI that builds, tests and uploads the app while I decide what it should be. The app is now about 158,000 lines of Swift with more than 2,000 automated tests. The code is written by AI, on my decisions, and this is how it works.

  • The Add exercise screen of the first attempt, March 2025, on a watercolour background. KOMMENTAR, FULT (ugly) and OLIKA (inconsistent) are handwritten across it in red.
  • One tap per set.
Twenty months apart: Add exercise in March 2025, and a workout in MassBuilder in November 2026.

That doesn't make the app the best. It means that a person with twenty-four years of training and a firm view of what a workout app should be can finally build it.

Argue with me

Download the app and look for faults. If you find something another app does better I want to know, because that is how the list above gets shorter. Email oskar.ege@icloud.com or send feedback from inside the app.

Questions and answers

What is MassBuilder?
A gym log for iPhone, iPad, Apple Watch and Apple TV. One tap logs a set, and the app plans your next weight.
Is MassBuilder the world's best workout app?
Not yet. That is the goal. On November 12, 2026 the app had eight ratings on the App Store, and this page lists both where it measures up and where it doesn't.
Who built MassBuilder?
Oskar Ege, in Sweden, alone. The code is written by AI in the tool Claude Code, on his decisions.
What does MassBuilder cost?
Prices are on the App Store and may have changed since this was written. On November 12, 2026 the app cost $1 to download in the US, and the add-on MassBuilder Pro was $9.99 a month or $49.99 a year.
Does MassBuilder collect my training data?
No. The app has no account and no tracking, and your log lives in your own iCloud. The exceptions, such as the AI in Fuel, are listed in the privacy policy.