Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Claude is not your architect. Stop letting it pretend

hollandtech.net

Post reply on HN
Print view View original

191–200 of 209 posts

Re: Claude is not your architect. Stop letting it pretend

#191

Post by mizzao »

himata4113 wrote: post #110

I have a good story to share that I came across recently. Around 2 years ago I had to clean up a mess because someone who doesn't really know what they're doing designed an instancing system for a game. They heavily used AI to design every part of it and it was awful. Data corruption, performance problems, lost items, race conditions everything you can think of was an issue. It took me 2 weeks just to get it to an "a…

2 years ago AI coding agents had barely emerged yet, right? Couldn't have gotten good results even if a good engineer was forced to use it...

Re: Claude is not your architect. Stop letting it pretend

#192

Post by operatingthetan »

gchamonlive wrote: post #164
operatingthetan wrote: post #124

Earlier quoted context omitted.

The machine in your scenario is just relaying human intent.

And what's the difference between that machine and LLMs?

You are using poor (inverse) analogies and don't seem to be actually addressing anthropomorphism.

Re: Claude is not your architect. Stop letting it pretend

#193

Post by sevenzero »

kgeist wrote: post #70
sevenzero wrote: post #28

Earlier quoted context omitted.

>Agents are way more than that, today. Not really though. They just iterate more and more.

Isn't that how many people program too? I remember some idea or pattern from previous projects, or something I read about on the internet. Then I code it in the most straighforward way, whatever comes to mind first. Then I sit back and analyze: does it look good architecturally? Do I like it? Does it even compile? Then I rewrite some parts to make it more sound. Rinse and repeat, until I'm satisfied. I usually don't…

Its absolutely valid, my point was just about the "agents are way more than that" part which simply isnt true.

I try to not re-iterate too much, but maybe thats due to me not wanting to work and working for a startup so time and motivation are hard to find.

Re: Claude is not your architect. Stop letting it pretend

#194

Post by Npovview »

AndrewKemendo wrote: post #176
Npovview wrote: post #140

Earlier quoted context omitted.

Do you use skills like superpowers and spec-kit in your teachings ?

No, I don’t know what those are. (Looked them up and I don’t teach every possible handler, but I teach people how to do structured inputs etc..) I teach TDD philosophy as well as conways law, parnas hiding etc…without using those terms So things like problem decomposition into tractable chunks minimum viable product, prototyping, how do you iterate, write the smallest possible test… you know things like this which ar…

Take it as you may but I got good results out of those. I would suggest that you take task like porting a code from c++ to go or equivalent complex task and use those skills along with your traditional knowledge of what you expect from a port and see how those skills do. Just annecdote from my side. You can do experimentation on your side.

Re: Claude is not your architect. Stop letting it pretend

#195

Post by jjav »

otar wrote: post #123
himata4113 wrote: post #110

I have a good story to share that I came across recently. Around 2 years ago I had to clean up a mess because someone who doesn't really know what they're doing designed an instancing system for a game. They heavily used AI to design every part of it and it was awful. Data corruption, performance problems, lost items, race conditions everything you can think of was an issue. It took me 2 weeks just to get it to an "a…

> AI is only as good as the person using it, that's why we have such vast range of what people "claim" AI can do and why everyone has way different opinions of it. Banger statement.

> > AI is only as good as the person using it, that's why we have such vast range of what people "claim" AI can do and why everyone has way different opinions of it.

> Banger statement.

True, but it's even a bit more removed than that.

For a version 0.9 prototype, AI can produce pretty awesome-feeling output even in the hands of the most incompetent. So it looks like a miracle that does it all, with zero effort.

It's later down the road that it starts hitting walls that need a competent architect in charge.

Re: Claude is not your architect. Stop letting it pretend

#197

Post by sph »

otar wrote: post #123
himata4113 wrote: post #110

I have a good story to share that I came across recently. Around 2 years ago I had to clean up a mess because someone who doesn't really know what they're doing designed an instancing system for a game. They heavily used AI to design every part of it and it was awful. Data corruption, performance problems, lost items, race conditions everything you can think of was an issue. It took me 2 weeks just to get it to an "a…

> AI is only as good as the person using it, that's why we have such vast range of what people "claim" AI can do and why everyone has way different opinions of it. Banger statement.

Shades of that famous quote by Kurt von Hammerstein-Equord.

What is terrifying is that those that claim 100x increased productivity are those with the least wisdom to tell good code from bad. They are the ones inundating the world with utter slop.

Re: Claude is not your architect. Stop letting it pretend

#198

Post by spacedcowboy »

HDThoreaun wrote: post #115
spacedcowboy wrote: post #20

Earlier quoted context omitted.

It actually got quite snippy with me, when I was trying to get it to debug some issues. It kept on saying that the "burn rate" wasn't progressing and "we" should refocus our efforts somewhere else. Eventually I got something like "I have told you three times now that this is not the best approach to be taking to reduce the burn-rate and you have not taken that advice". And it stopped helping out. So I was blunt, and…

Dont argue with LLMs. Sometimes they lose the plot, when that happens simply flush the context and start over.

Reframing its priorities worked well in this case - just telling it that the metrics it was using weren't appropriate made it reset its world-view.

Re: Claude is not your architect. Stop letting it pretend

#199

Post by Npovview »

AndrewKemendo wrote: post #176
Npovview wrote: post #140

Earlier quoted context omitted.

Do you use skills like superpowers and spec-kit in your teachings ?

No, I don’t know what those are. (Looked them up and I don’t teach every possible handler, but I teach people how to do structured inputs etc..) I teach TDD philosophy as well as conways law, parnas hiding etc…without using those terms So things like problem decomposition into tractable chunks minimum viable product, prototyping, how do you iterate, write the smallest possible test… you know things like this which ar…

Follow these channels or videos for better understanding of skills and how they are using it.

https://www.youtube.com/@mattpocockuk

https://www.youtube.com/@EricWTech

https://www.youtube.com/watch?v=eRS3CmvrOvA

Re: Claude is not your architect. Stop letting it pretend

#200

Post by himata4113 »

mizzao wrote: post #191
himata4113 wrote: post #110

I have a good story to share that I came across recently. Around 2 years ago I had to clean up a mess because someone who doesn't really know what they're doing designed an instancing system for a game. They heavily used AI to design every part of it and it was awful. Data corruption, performance problems, lost items, race conditions everything you can think of was an issue. It took me 2 weeks just to get it to an "a…

2 years ago AI coding agents had barely emerged yet, right? Couldn't have gotten good results even if a good engineer was forced to use it...

People in the know already had coding agents, they just needed a lot more back and forward as they weren't trained to be "agentic" and they were mostly cli based... then claude code was born because anthropic saw the potencial. The person in question used warp.dev which happened to bundle AI into so they stumbled into it more than anything.
Post reply on HN