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

Claude Code Unpacked : A visual guide

ccunpacked.dev

Post reply on HN
Print view View original

171–180 of 464 posts

Re: Claude Code Unpacked : A visual guide

#171

Post by hombre_fatal »

troupo wrote: post #135
fragmede wrote: post #129

Earlier quoted context omitted.

How many LoC should it be, for that kind of program?

It's a TUI API wrapper with a few commands bolted on. I doubt it needs to be more than 20-50kloc. You can create a full 3D game with a custom 3D engine in 500k lines. What the hell is Claude Code doing?

Software doesn’t end at the 20k loc proof of concept though.

What every developer learns during their “psh i could build that” weekendware attempt is that there is infinite polish to be had, and that their 20k loc PoC was That said, doesn't TFA show you what they use their loc for?

Re: Claude Code Unpacked : A visual guide

#172

Post by troupo »

spiderfarmer wrote: post #143
troupo wrote: post #135

Earlier quoted context omitted.

It's a TUI API wrapper with a few commands bolted on. I doubt it needs to be more than 20-50kloc. You can create a full 3D game with a custom 3D engine in 500k lines. What the hell is Claude Code doing?

Comments like these remind me of the football spectators that shout "Even I could have scored that one" when they see a failed attempt. Sure. You could have. But you're not the one playing football in the Champions League. There were many roads that could have gotten you to the Champions League. But now you're in no position to judge the people who got there in the end and how they did it. Or you can, but whatever.

> Sure. You could have. But you're not the one playing football in the Champions League.

The only reason people are using Claude Code is because it's the only way to use their (heavily subsidized) subscription plans. People who are okay with using and paying for their APIs often opt out for other, better, tools.

Also, analogies don't work. As we know for a fact that Claude Code is a bloated mess that these "champions league-level engineers" can't fix. They literally talk about it themselves: https://news.ycombinator.com/item?id=47598488 (they had to bring in actual Champions League engineers from bun to fix some of their mess).

Re: Claude Code Unpacked : A visual guide

#173

Post by ernst_klim »

fragmede wrote: post #129
ernst_klim wrote: post #123

> 500k lines of code Isn't it a simple REPL with some tools and integrations, written in a very high level language? How the hell is it so big? Is it because it's vibecoded and LLMs strive for bloat, or is it meaningful complexity?

How many LoC should it be, for that kind of program?

Well FFmpeg is roughly 1500k, but it's C+Asm and it's dozens of codecs and pretty complex features. SBCL is around 500k I guess.

I'm not saying that this is necessarily too much, I'm genuinely asking if this is a bloat or if it's justified.

Re: Claude Code Unpacked : A visual guide

#174

Post by __s »

throwatdem12311 wrote: post #170

I know it seems counter-intuitive but are there any agent harnesses that aren’t written with AI? All these half a million LoC codebases seem insane to me when I run my business on a full-stack web application that’s like 50k lines of code and my MvP was like 10k. These are just TUIs that call a model endpoint with some shell-out commands. These things have only been around in time measured in months, half a million L…

Check out pi coding agent, https://pi.dev

Re: Claude Code Unpacked : A visual guide

#175

Post by sunir »

amangsingh wrote: post #148

A 500k line codebase for an agent CLI proves one thing: making a probabilistic LLM behave deterministically is a massive state-management nightmare. Right now, they're great for prompting simple sites/platforms but they break at large enterprise repos. If you don't have a rigid, external state machine governing the workflow, you have to brute-force reliability. That codebase bloat is likely 90% defensive programming;…

It’s not surprising. There has been quite a bit of industrial research in how to manage mere apes to be deterministic with huge software control systems, and they are an unruly bunch I assure you.

Re: Claude Code Unpacked : A visual guide

#176

Post by p-e-w »

amangsingh wrote: post #148

A 500k line codebase for an agent CLI proves one thing: making a probabilistic LLM behave deterministically is a massive state-management nightmare. Right now, they're great for prompting simple sites/platforms but they break at large enterprise repos. If you don't have a rigid, external state machine governing the workflow, you have to brute-force reliability. That codebase bloat is likely 90% defensive programming;…

> A 500k line codebase for an agent CLI proves one thing: making a probabilistic LLM behave deterministically is a massive state-management nightmare.

Considering what the entire system ends up being capable of, 500k lines is about 0.001% of what I would have expected something like that to require 10 years ago.

You can combine that with all the training and inference code, and at the end of the day, a system that literally writes code ends up being smaller than the LibreOffice codebase.

It boggles the mind, really.

Re: Claude Code Unpacked : A visual guide

#177

Post by shuntaka9576 »

A year ago I wouldn't have guessed a TUI could be a competitive advantage. But "harness engineering" became a thing, and it turns out the agent wrapper — tool orchestration, context management, permission flows — is where real product value lives. Not as much as the models themselves, but more than most people expected. This leak is a painful reminder of that.

Re: Claude Code Unpacked : A visual guide

#178

Post by dolomo »

amangsingh wrote: post #148

A 500k line codebase for an agent CLI proves one thing: making a probabilistic LLM behave deterministically is a massive state-management nightmare. Right now, they're great for prompting simple sites/platforms but they break at large enterprise repos. If you don't have a rigid, external state machine governing the workflow, you have to brute-force reliability. That codebase bloat is likely 90% defensive programming;…

[flagged]

Re: Claude Code Unpacked : A visual guide

#179

Post by kristopolous »

kristopolous wrote: post #106
spondyl wrote: post #97

Earlier quoted context omitted.

Well, I assume this is all just generated with Claude Code, right? Whether there is much back and forth with the LLM is a valid question and nothing wrong with generating websites (I do it too for some side projects). Claude loves generating websites with a particular style of serif font. We also saw this with https://tboteproject.com/timeline/ and I've just generally seen it from various designs that coworkers have…

I really wish I had that clout-chasing gene - it doesn't even occur to me until I see someone else do it. I'm serious. The hype chasing clearly clearly matters. . things like this: https://github.com/instructkr/claw-code I mean ok, serious people put in years of effort for 100 of those stars ... it's continually wild how extremely irrelevant hard effortful careful work is. I think that's the game. Get up, look at the…

That was the leaked code and now it's just some random dudes harness btw. He swapped it out. Did a sloppy find and replace for "claude" and made it claw.

It's sloppy work

Does not matter. Sloppiness is unimportant

Re: Claude Code Unpacked : A visual guide

#180

Post by bogdanoff_2 »

amangsingh wrote: post #148

A 500k line codebase for an agent CLI proves one thing: making a probabilistic LLM behave deterministically is a massive state-management nightmare. Right now, they're great for prompting simple sites/platforms but they break at large enterprise repos. If you don't have a rigid, external state machine governing the workflow, you have to brute-force reliability. That codebase bloat is likely 90% defensive programming;…

What do you mean by "actually governing the agents at the system level", and how is it different from "herding cats"?
Post reply on HN