I don't care how the code got into your IDE but I want you to care

▶

I don't care how the code got into your IDE.

AI as a New Way of Programming

I just want to get to the conclusion, which is precisely this: AI, yes, of course. It's perfectly fine, because the more efficient we are and the better we can do the same work—even better and in less time—the better. But, again, AI is always a responsible assistant.

That is, not everything it spits out may be what you're looking for: it may not be functional, it may be outdated code. For example, often with packages when we're installing, it shows you an old implementation. Therefore, it's your responsibility to always keep that in mind.

Three key points

1. It's not enough to accept what a model produces. I think it's pretty obvious, but based on what I just said, you always have to review it, monitor it, adjust it, and make sure it respects the conventions of the project you're following.
   1. For example, if you put all your variables in lowercase and with underscores to separate them, but the model generates everything in uppercase, it won't fit. You have to adjust it so it looks like part of the project and not a Frankenstein.
2. Maintainability is key. We always have to generate code that is maintainable, scalable, and reusable. You know what I mean: it's one of the conventions I follow here for my Django course and also for the online store.
   1. These are projects I'm trying to keep more of a middle ground for developers. Unlike the rest of my courses, where the focus is on explaining how the technology works, in these cases I focus on making the project fully functional and scalable. That's why, for example, for these projects we created a layer-based system, which I summarize in this video in case you're interested, specifically on the topic of payments.
3. Quality matters more than speed. Software isn't ephemeral prototypes, but systems that must be sustainable over time. You have to adjust it based on your conventions—assuming they're good practices—and ensure it's maintainable.

Conclusion

Basically, the bottom line is this: software quality and sustainability come before mere speed. What do you think about all this? I'll read your comments, and we'll see you in another video.