3 ms·

I find their value depends a lot on what I'm doing. Anything easy I'll get insane leverage, no exaggeration I'll slap together that shit 25x faster. It's seen

I find their value depends a lot on what I'm doing. Anything easy I'll get insane leverage, no exaggeration I'll slap together that shit 25x faster. It's seen likely billions of lines of simple CRUD endpoints, so yeah it'll write those flawlessly for you. Anything the difficult or complex, and it's really a coinflip if it's even an advantage, most of the time it's just distracting and giving irrelevant suggestions or bad textbook-style implementations intended to demonstrate a principle but with god-awful performance. Likely because there's simply not enough training data for these types of tasks. With this in mind, I don't think it's strange that junior devs would be gushing over this and senior devs would be raising a skeptical eyebrow. Both may be correct, depending on what you work on.

jeremy151 2y ago
I think for me, I'm still learning how to make these tools operate effectively. But even only a few months in, it has removed most all the annoying work and lets me concentrate on the stuff that I like. At this point, I'll often give it some context, tell it what to make and it spits out something relatively close. I look it over, call out like 10 things, each time it says "you're right to question..." and we do an iteration. After we're thru that, I tell it to write a comprehensive set of unit tests, it does that, most of them fail, it fixes them, and them we usually have something pretty solid. Once we have that base pattern, I can have it pattern and extend variants after the first solid bit of code. "Using this pattern for style and approach, make one that does XYZ instead." But what I really appreciate is, I don't have to do the plug and chug stuff. Those patterns are well defined, I'm more than happy to let the LLM do that and concentrate on steering whether it's making a wise conceptual or architectural choice. It really seems to act like a higher abstraction layer. But I think how the engineer uses the tool matters too.

jeremy151 2y ago

I think for me, I'm still learning how to make these tools operate effectively. But even only a few months in, it has removed most all the annoying work and lets me concentrate on the stuff that I like. At this point, I'll often give it some context, tell it what to make and it spits out something relatively close. I look it over, call out like 10 things, each time it says "you're right to question..." and we do an iteration. After we're thru that, I tell it to write a comprehensive set of unit tests, it does that, most of them fail, it fixes them, and them we usually have something pretty solid. Once we have that base pattern, I can have it pattern and extend variants after the first solid bit of code. "Using this pattern for style and approach, make one that does XYZ instead." But what I really appreciate is, I don't have to do the plug and chug stuff. Those patterns are well defined, I'm more than happy to let the LLM do that and concentrate on steering whether it's making a wise conceptual or architectural choice. It really seems to act like a higher abstraction layer. But I think how the engineer uses the tool matters too.