Frameworks aren't just human-serving abstractions - they're structural abstractions that allow for performant code, or even being able to achieve certain behaviours.
Sure, you could write a frontend without something like react, and create a backend without something like django, but the code generated by an LLM will become similarly convoluted and hard to maintain as if a human had written it.
LLM's are still _quite_ bad at writing maintainable code - even for themselves.
A similar thing I've noticed a trend of recently in frontend react codebases is overuse of memoization. It seems as though people don't realize how it works and that it is often _less_ performant than just doing some low cost computation on each render (like a comparison or basic math).
I've always liked an asymmetric layout when working with two monitors such that I have one "main" monitor in front used for my main task (e.g. my terminal for work) then a secondary to the side that is used for supporting windows - browser, music, pdfs, or another terminal depending on the work. It helps keep an established hierarchy, and you basically just use your neck as a window switcher.
The issue that many developers have with time tracking is that it requires you to either translate your work into management speak in order to appease the management, or write what you're actually doing and have tons and tons of small blocks in your time tracker that make you look like you're being unproductive and trying to obscure something. Both options add unnecessary mental overhead and interrupt any possibility of having a "flow state".
Writing code and working on software is not a linear process, so writing "worked on feature X" for your entire morning may be accurate, but it's not fine grained enough for what the managers want to see and it doesn't really serve any purpose at that point when you already have daily stand-ups and sprint meetings every week.
This likely isn't the case at all companies and with all managers, but I suspect many have the same experience I've described.
Sure, you could write a frontend without something like react, and create a backend without something like django, but the code generated by an LLM will become similarly convoluted and hard to maintain as if a human had written it.
LLM's are still _quite_ bad at writing maintainable code - even for themselves.