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

How Anthropic teams use Claude Code

anthropic.com

Post reply on HN
Print view View original

231–240 of 248 posts

Re: How Anthropic teams use Claude Code

#231

Post by godelski »

const_cast wrote: post #216
godelski wrote: post #191

Earlier quoted context omitted.

> We're still working out the edge cases The difficult part is that like with FSD, it's mostly edge cases

Driving is just mostly edge cases. I've thought about it a lot, but I think automating driving is much harder than automating even air travel. Sure the air is 3 dimensions, but driving is too dynamic and volatile. Every single road is different, and you have to rely on heuristics meant for humans. It's stupid easy for humans to tell what is a yellow line and what a stop sign looks like, but it's not so easy for compu…

  > I think automating driving is much harder than automating even air travel.
This is a pretty common belief. Well supported too since we've had a high level of automation in aviation for decades. Helps that things are very monitored. 3 dimensions provides a lot of benefits given that it makes for a lower density. Not to mention people don't tend to be walking around in the sky

Re: How Anthropic teams use Claude Code

#232

Post by godelski »

thunky wrote: post #226
godelski wrote: post #192

Earlier quoted context omitted.

> What makes you think that agents can't iterate? Please RTFA or RTF top most comment in the thread. Can they? Yes. Will they reliably? If so, why would it be better to restart... But the real answer to your question: personal experience

> Please RTFA TFA says: Engineers use Claude Code for rapid prototyping by enabling "auto-accept mode" (shift+tab) and setting up autonomous loops in which Claude writes code, runs tests, and iterates continuously. The tool rapidly prototypes features and iterates on ideas without getting bogged down in implementation details

Don't cherry-pick, act in good faith. I know you can also read the top comment I suggested.

I know it's a long article and the top comment is hard to find, so allow me to help

  > Treat it like a slot machine
  >
  > Save your state before letting Claude work, let it run for 30 minutes, then either accept the result or start fresh rather than trying to wrestle with corrections. ***Starting over often has a higher success rate than trying to fix Claude's mistakes.***
*YOU* might be able to iterate well with Claude but I really don't think a slot machine is consistent with the type of iteration we're discussing here. You can figure out what things mean in context or you can keep intentionally misinterpreting. At least the LLM isn't intentionally misinterpreting

Re: How Anthropic teams use Claude Code

#233

Post by tayo42 »

thunky wrote: post #230
tayo42 wrote: post #229

Earlier quoted context omitted.

Appending more context to the existing prompt means it's a different prompt still... The text isn't the same

I'm not sure what your point is? Think of it like an append only journal. To correct an entry you add a new one with the correction. The LLM sees the mistake and the correction. That's no worse then mutating the history.

thats not how it works.

You put in its context window some more information, then roll the dice again. And it produces text again token by token. Its still not planning ahead and its not looking back at what was just generated. There's no guarantee everything stays the same except the mistake. This is different then editing in place. you are rolling the dice again.

Re: How Anthropic teams use Claude Code

#234

Post by moomoo11 »

AstroBen wrote: post #179
moomoo11 wrote: post #134

Earlier quoted context omitted.

Not really. Is it really value add to my life that I know some detail on page A or have some API memorized? I’d rather we be putting smart people in charge of using AI to build out great products. It should make things 10000x more competitive. I’m for one excited AF for what the future holds. If people want to be purists and pat themselves on the back sure. I mean people have hobbies like arts.

> Is it really value add to my life that I know some detail on page A or have some API memorized? yes, actually. Maybe not intimate details but knowing what's available in the API heavily helps with problem solving

Right I’m saying I know them and where to find them. But my brain space is better occupied learning more things that exist than some mundane crap that I can delegate out.

Re: How Anthropic teams use Claude Code

#237

Post by oblio »

lumost wrote: post #154
oblio wrote: post #60

Earlier quoted context omitted.

$200h * 8 * 5 * 4 * 12 = $384 000 per year. You're like in the top 0.05% of earners in the software field. Of course, if you save 10 hours per month, the math starts making more sense for others. And this is assuming LLM prices are stable, which I very much doubt they are, since everyone is price dumping to get market share.

The percentages in the field are skewed, FAANG employ a vast number of engineers.

No, they don't. FAANG probably employs 400 000 programmers worldwide, and I think the US alone probably has about 3-4 million programmers. Worldwide there are probably 30 million.

And even for FAANG, an SDE for them in Spain makes 60-100k total comp, not 400k.

Re: How Anthropic teams use Claude Code

#238

Post by withinboredom »

david38 wrote: post #221
withinboredom wrote: post #104

Earlier quoted context omitted.

At least when you tell a human the indentation is wrong, they can fix it on the first try. Watched an AI agent last night try to fix indentation by using sed for 20 minutes before I just fixed it myself after cringing.

Why the hell would anyone do this instead of using any one of dozens of purpose written tools that accept configuration files? They take less than a second to run, can run on every save, and are free

My point exactly...

Re: How Anthropic teams use Claude Code

#239

Post by nojito »

godelski wrote: post #232
thunky wrote: post #226

Earlier quoted context omitted.

> Please RTFA TFA says: Engineers use Claude Code for rapid prototyping by enabling "auto-accept mode" (shift+tab) and setting up autonomous loops in which Claude writes code, runs tests, and iterates continuously. The tool rapidly prototypes features and iterates on ideas without getting bogged down in implementation details

Don't cherry-pick, act in good faith. I know you can also read the top comment I suggested. I know it's a long article and the top comment is hard to find, so allow me to help > Treat it like a slot machine > > Save your state before letting Claude work, let it run for 30 minutes, then either accept the result or start fresh rather than trying to wrestle with corrections. ***Starting over often has a higher success r…

That’s actually an old workflow. Nowadays you spin up a thin container and let it go wild. If it messes up you simply just destroy the container, undo the git history and try again.

Takes no time at all.

Re: How Anthropic teams use Claude Code

#240

Post by nojito »

danielbln wrote: post #210
AstroBen wrote: post #188

Earlier quoted context omitted.

How do you know the test suite is comprehensive if you don't have a grasp on what's happening? Not a gotcha I'm just extremely skeptical that AI is at a point to have the level of responsibility you're describing and have it turn into good code long term

I do TDD. Plan out the feature, then have it write test. Scan the tests, check if everything is kosher. Then let it do the implementation, rinse and repeat.

How do you stop it from changing or modifying the test to get to pass?
Post reply on HN