I studied management. Not computer science, not software engineering. Tech always interested me, but for most of my education that stayed abstract. I could follow it, read about it, get genuinely interested in it. I just couldn’t build anything.
I joined SAP’s Innovation Labs as an intern in 2019, and that’s where the opportunity came up. The Labs is a good place to experiment and try things. But I kept finding myself in situations where I had an idea I couldn’t act on myself. I could describe it, sketch it out, talk it through, but making it was someone else’s job.
Starting from scratch
I had some basic HTML, CSS, and JavaScript to build on. Over the following months I put together landing pages for real customer opportunities at work, about 20 of them in total. The running joke about web development is that nobody ever truly masters centring a div, and I can confirm that. The basics stopped needing much thought though.
Then Python and Node.js. Raspberry Pis and Arduinos. 3D printed things, connected a robot arm to an AI chatbot, small automations around the flat. The physical stuff was satisfying in a different way. A Python script triggering something real in the world feels more concrete than a web page.
The bootcamp
The thing that pushed me furthest was Angela Yu’s iOS development bootcamp on Udemy. It covers a lot of ground and takes you seriously as a learner. I worked through it over several months, getting comfortable with Swift, learning ARKit and CoreML.
But courses have a ceiling. At some point you have to build something real, with actual constraints, actual edge cases, something that has to survive contact with people who aren’t you.
Shipping something
The forcing function I gave myself was the App Store. Apple’s review process is not lenient. An app has to actually work, handle edge cases, not crash. You can’t fake it like you can with a demo.
That became Colour Kit, an iOS app for creating and discovering colour palettes and visualising them on real surfaces using AR. Swift, ARKit, CoreML, Firebase. I built it during lockdown.
It shipped in September 2020. Getting the approval email was mostly relief. The app is simple. But it was a real thing on a real platform that worked. That’s different from a demo. A demo can be propped up. A shipped product either works or it doesn’t.
Where I ended up
Still learning. What I didn’t expect is how much actually building things changes how you read the bigger picture. Once you have a feel for where real complexity lives and what the constraints are, you notice when things are being oversimplified.
The management background wasn’t wasted. But it only gets you so far when you can’t also make things.
Two years later, I mostly can.