Is that possible to build an iOS app with ChatGPT?

As developers, we at Xuntos are always curious about all development opportunities. You try out new developments and code creative new applications. With ChatGPT in particular, the possibilities are endless. In this blog, our senior developer Jan Willem Bergsma answers the question: Is that possible to build an iOS app with ChatGPT?

How I Built My First iOS App with ChatGPT

Blog by Jan Willem Bergsma senior developer

With more than 25 years of experience as a programmer, I still enjoy discovering new technologies and programming languages. At home, I drive them crazy because I automate almost everything — from smart thermostats and a self-programmed weather station to automatically opening the gate and triggering the alarm (including accidentally setting it off 🙃). These solutions not only offer me daily convenience, but also the perfect opportunity to keep experimenting with the latest technical innovations.

Over the past few years, I've increasingly started using ChatGPT to solve my programming problems. If I get stuck in my code, I ask ChatGPT my question. Surprisingly, nine times out of ten, I get a working solution. That saves me a lot of time and makes me a lot more efficient as a programmer. I also use ChatGPT for code review, documentation, and sometimes even inspiration when writing complex functions.

From idea to working ios app

Because ChatGPT helps me so much with smaller programming languages, I decided to go one step further: build a complete iOS app using ChatGPT. To do this, I chose Swift, Apple's programming language, and the Xcode platform.

The goal - What will ChatGPT and I build?

Build an application that integrates with my Home Assistant system. For example, I wanted my stove to turn on automatically when I get home, or for my gate to open when I arrive. While there are already existing apps that offer these kinds of features, I often found them not accurate enough. So I decided to build something myself.

What exactly did the app need to be able to do?

The process started with writing out my functional requirements: what exactly did the app need to be able to do? I entered these requirements into ChatGPT and within seconds I got an entire Xcode project back. Of course, there were a few minor errors in the code, but after two or three adjustments, the project worked flawlessly.

After that, I continued to refine the app. I added new screens, checked the functionalities, and optimized the code where necessary.

What lessons did I learn?

  • Chat GPT can get stuck in a loop
    Although Chat GPT is impressive, I noticed that it sometimes starts talking in circles when it comes to complex problems. It then gave the same solution over and over, even if it didn't work. By diving into the documentation myself or trying a different angle, I was still able to quickly come to a solution. It is therefore important to keep thinking for yourself and not to think fully
  • How easily Chat GPT deals with error messages
    lorem ipsum and more text
  • Chat GPT can get stuck in a loop
    While ChatGPT is impressive, I noticed that it sometimes starts talking in circles when it comes to complex problems. It then gave the same solution over and over, even if it didn't work. By diving into the documentation myself or trying a different angle, I was still able to quickly come to a solution. So it's important to keep thinking for yourself and not become completely dependent on AI.
  • How easily Chat GPT deals with error messages
    By uploading a screenshot of an error message and asking for help, I often immediately got a useful solution. This makes debugging much faster and more efficient.

The end result

My app is now ready and almost available in the App Store. The accompanying website, including the logo and documentation, was also built entirely with AI tools. The website can be viewed at www.simplyanapp.nl Everything, from the app to the branding, has been generated by AI.

90% of the app was built by ChatGPT and
10% was added by thinking logically and adjusting small details

Reflection

Building this app using ChatGPT has shown me how powerful AI already is. It not only speeds up the development process, but also lowers the threshold for trying out new ideas. Although there is still room for improvement, I think we will have to program less and less manually in the near future.

For now, I'll keep experimenting with ChatGPT and other AI tools. Who knows what the future holds, but one thing is certain: programming will never be the same again.

Want to know more about developing with AI?

Xuntos is increasingly developing and experimenting with AI or ChatGPT.
What are we working on? Or what products we develop based on AI?
Continue reading AI & Web development for answers to these questions.

Or ask about the options by contact to record with us.

Getting to know

Heb je vragen? Of wil je samen sparren?

We denken graag met je mee hoe jouw organisatie digitaal kan versnellen. Of het nu gaat om een UX, technische uitdaging of AI oplossingen. Stuur een berichtje en we nemen zsm contact met je op.