Vibe coding / AI code generation for Qt based applications

Axel Spoerl

Dear users,

which experiences have you made regarding AI code generation for applications based on Qt?
Do you have any thoughts, opinions, examples to share?
Excited to read and discuss with you!

Have a nice day
Axel

Mizmas

@Axel-Spoerl I just use ChatGPT as documentation instead of googling for functions/flags

SimonSchroeder

I'm not really using AI for programming (yet). Still, I do follow the discussions about this. There is horror stories about the AI "knowingly" deleting databases even though it was explicitly told not to. At least automatic AI agents seem to be to dangerous to use for software that is supposed to make money.

Recent comments about this topic also talk about AI making seemingly good changes when looking at individual commits. But once you zoom out and look at the full picture it might become a total maintenance nightmare. Our software already has a lot of technical dept and I'm not willing to increase that.

Finally, I'm not sure if our software is too large for current LLMs to digest entirely. Making choices about changing the software architecture without having the full picture does not seem like sound advice to me. At least I would wait for even larger context windows for AI to work on our software.

The youtuber Primeagen also said that he stopped using AI coding plugins in his code editor. His reason: He would start typing a few letters and then wait a few seconds for the AI to autocomplete the next couple of lines. In the mean time he could have written more code himself (not necessarily faster than AI, but also not much slower than AI). I personally believe that it is important to write code yourself to gain understand for the software you are working on. If AI is doing the work you are turning your brain off. In a couple of years it will be hard to bounce back from badly AI-written code as nobody will understand the source code anymore and there will be very few programmers out there that will still be able to work throught the mess the AI created.

Maybe I am a bit too negative about AI. Also, I enjoy programming way to much to be degraded to be only reviewing AI-written code. I'm supposed to do the fun part–not the AI. Does anybody have a more positive take on this?

cristian-adam

@Axel-Spoerl I've created my own local LLM Qt Creator plugin using AI. See https://forum.qt.io/topic/163001/local-llm-assisted-text-completion-for-qt-creator for details.

KH-219Design

Like @SimonSchroeder , I'm not really using AI for programming (yet), but I, too, follow the discussions about this (sometimes).

Something I have been pondering recently, without coming to any conclusions...

Over the years, most programmers/coders/engineers that I have worked with are much more comfortable writing code than sitting down and intently reading through it methodically and critically (especially for "novel" code for which the reader was not also the author).

I have regularly given my favorite "pet" advice to people just starting out that they should try to get really comfortable and really fluent in reading and analyzing code, and practice that skill as often as they practice writing code.

So....

In a hypothetical world where the AI writes 100% of the code, then either it will somehow be "perfectly coded" and need no review (which I cannot imagine, since the "spec" could have errors), or....

the AI writes 100% of the code and it falls upon us to READ all the code, which is something a lot of us do poorly.

This simultaneously brings up two competing feelings for me:

excitement/hope that code reading will now become more recognized as the highly-prized skill that I always believed it to be.
pessimism/dread that nobody will be reading all the "AI slop" code that is potentially about to be dumped into the world's code repositories, and that there already aren't enough good readers/reviewers to deal with it all