Comment by Nextgrid
11 hours ago
This is also missing step 6: when your solution breaks in production or needs to be extended.
Having done step 2 & 3 by hand is the difference between being able to fix/extend it quickly with no further damage or fumbling around like an idiot and sometimes breaking more stuff in the process.
It's been pretty disheartening to watch SWEs rediscover/reinvent the SDLC.
The most demoralising part of using LLMs is that now I spend most of my day on SDLC - even hobby projects.
>Having done step 2 & 3 by hand is the difference between being able to fix/extend it quickly with no further damage or fumbling around like an idiot and sometimes breaking more stuff in the process.
Plenty of shitty spaghetti code has been written by human hands.
LLMs can write good, maintainable code too, but they need to be kept on a shorter leash with focused goals.
I'm not talking about code quality (though indeed that could be a problem too), I'm talking about understanding the code - having written the code by hand means you don't need to rediscover it from scratch.
(with human-written code I can reach out to the person who wrote it and let them deal with it, and they will have the understanding of said code, even if it is bad by quality measures. With LLMs there is nobody who understands said code, regardless of its quality)
To be fair though:
"That code is horrible! Who the hell wrote this?!"
git blames
sees their own name
Isn't that an experience everybody makes? For me it was like a year into working professionally. And I feel it's part of being a professional to know that you're not going to remember it in the long term and therefore provide proper documentation of why you did what.
1 reply →
>having written the code by hand means you don't need to rediscover it from scratch.
Hard disagree. If you're in a small-to-medium size codebase every day, then sure, but I have a couple personal projects that are complex enough that even though I wrote them before LLMs, I still need to go back and discover/fumble around before I can confidently make changes.
I have definitely looked at my own hand-written code and had a "wtf does this even do" moment.
While that might be true, the more one equips LLMs for building the less one will be capable of keeping LLMs “on a shorter leash with focused goals” in the long run