It is one of the game changers of AI.
It used to be a that the barrier of entry of creating a new tool was high - so we had to use popular pieces of software, often stretching them, or writing plugins (that had their own constraints).
Now it is often easier to write from scratch a new piece of software, for which with have full control.
We can stand on the shoulders of giants - not just "a giant".
To me, this article demonstrated the value of knowing your domain and its particular constraints, as well as a good understanding of prior art. Together with the only 1000 line demo it took to produce (which I suppose could have been aided by an LLM), I did not finish reading with any specific appreciation for AI.
No LLMs were used, at least not in any substantial way.
I'm sorry, I don't get your comment. What makes you think the approach or the code described in the article were developed by an LLM?
I don't know how the code was created.
But based on my experience with multiple project, both current frameworks and AI changed the game. I used to much more reply on existing software and cursing that they don't do what I want - as the time to create a personalized tool was simply much to afford. Now often it is quicker to create a personalized tool than fight with existing one.
Probably, but I see no connection with the article we're discussing.