Has AI made you a better problem solver, or just faster at finding solutions?
Something I've noticed since I started building with AI is that the way I approach problems has completely changed.
When I first started, I'd describe what I wanted, let AI generate the code, and hope everything worked lol.
Now I find myself spending more time thinking about why something should work a certain way, what could go wrong, and how changing one thing might affect everything else.
It's interesting because I started with basically zero coding experience, but I've learned a surprising amount about software just from trying to understand and troubleshoot what AI builds.
At the same time, I wonder how much of that knowledge I'd actually retain if the AI disappeared tomorrow.
For everyone building with AI, do you feel like you're genuinely becoming a better developer, or just getting better at working with AI?
And do you think there's even a meaningful difference between the two anymore?
p/reso
p/runevr
p/producthunt
p/google-cloud-platform
p/many-displays
p/slotify-2
p/airuncode
p/ai-context-flow
p/kodlo
p/aperture-7
p/lasso-11
p/raycast
p/dino-kids
p/kuc-kuc
p/supabase
p/jev
p/offiai
p/murror
p/dex-by-exmergo
p/htmldrag
Replies
for me the stuff that stuck came from the failures.. two claude code sessions in one repo and one killed the other's run because it matched a process by name. i'd never have learned to kill the exact pid you started if that hadn't bitten me. so maybe the difference is whether you go read why it broke or just ask it to fix it again
For me it feels more like reach than skill. I've done everything on my own for years, from development to marketing to tax and legal, and working with agents feels like all of that got stretched a lot further. The thinking about why something should work is still mine, though.
Definitely becoming a better developer if you DON'T trust your coding AI and treat it like the stupidest guy in your company. 😂
Watching it come up with overcomplicated, bloated solutions often teaches you what's NOT necessary at all. But understanding why it reasoned that way also opens up perspectives you might never have discovered on your own as well. It's a bit like learning from a crazy person who actually has some substance (the training data from millions of repositories and developers).
At the end of the day, AI is full of assumptions and biases that constantly collide with reality. And when you experience that often enough, you start using your own brain more. You learn to break complex workflows down into simple, individual tasks and prompt the AI to create the simplest, dumbest pieces of code possible. Pieces that, when connected the right way, become something really smart, even though none of them knows s*** on its own.
I think I'm only as good a developer as I am today because I've spent so many hours working with this crazy, stupid guy. I've learned both how NOT to do it and how to do it.
It's a weird relationship between humans and AI: two neural networks, different substrates, interacting in unpredictable ways.
It reminds me of my puppy and me. You never really know who's the boss. Am I, because I hold the leash? Or is he, because I follow the leash? 😂