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

121–130 of 248 posts

Re: How Anthropic teams use Claude Code

#121

Post by a_bonobo »

godelski wrote: post #76
throwmeaway222 wrote: post #8

Earlier quoted context omitted.

Thanks for the tip - we employees should run and re-run the code generation hundreds of times even if the changes are pretty good. That way, the brass will see a huge bill without many actual commits. Sorry boss, it looks like we need to hire more software engineers since the AI route still isn't mathing.

Unironically this can actually be a good idea. Instead of "rerunning," run in parallel. Then pick the best solution. Pros: - Saved Time! - Scalable! - Big Bill? Cons: - Big Bill - AI written code

This repo has a pattern where the in parallel jobs have different personalities: https://github.com/tokenbender/agent-guides/blob/main/claude...

Re: How Anthropic teams use Claude Code

#122

Post by alwillis »

oblio wrote: post #78
CMCDragonkai wrote: post #65

Earlier quoted context omitted.

Across most anglosphere countries and tech cities - wages and salaries far outstrip what you can get for AI. AI is already objectively cheaper than human talent in rich countries. Is it as good? Yea I'd say it's better than most mid to junior engineers. Can it run entirely by itself? No, it still needs HITL.

Again, those prices aren't stable. Nobody is investing half a trillion in a tech without expecting a 10x return. And fairly sure soon those $20/month subscriptions will sell your data, shove ads everywhere AND basically only allow you to get that junior dev for 30 minutes per day or 2 days a month. And the $200/month will probably be $500-1000 with more limitations. Still cheap, but AI can't run an entire project, ca…

What’s different is all the open weight models like Kimi-k2 or Qwen-3 Coder that are as good and, depending on the task, better than Anthropic’s Sonnet model for 80% less via openrouter [1] and other similar services.

You can use these models through Claude Code; I do it everyday.

Some developers are running smaller versions of these LLMs on their own hardware, paying no one.

So I don’t think Anthropic and the other companies can dramatically increase their prices without losing the customers that helped them go from $0 to $4 billion in revenue in 3 years.

Users can easily move between different AI platforms with no lock-in, which makes it harder to increase prices and proceed to enshitify their platforms.

[1]: https://openrouter.ai/

Re: How Anthropic teams use Claude Code

#123

Post by godelski »

DANmode wrote: post #99
godelski wrote: post #76

Earlier quoted context omitted.

Unironically this can actually be a good idea. Instead of "rerunning," run in parallel. Then pick the best solution. Pros: - Saved Time! - Scalable! - Big Bill? Cons: - Big Bill - AI written code

Have you seen human-written code?

I've taught undergraduates and graduates how to code. I've contributed to Open Source projects. I'm a researcher and write research code with other people who write research code.

You could say I've seen A LOT of poorly written human generated code.

Yet, I still trust it more. Why? Well one of the big reasons is exactly what we're joking about. I can trust a human to iterate. Lack of iteration would be fine if everything was containerized and code operates in an unchanging environment[0]. But in the real world, code needs to be iterated on, constantly. Good code doesn't exist. If it does exist, it doesn't stay good for long.

Another major problem is that AI generates code that optimizes for human preference, not correctness. Even the terrible students who were just doing enough to scrape by weren't trying to mask mistakes[1], but were still optimizing for correctness, even if it was the bare minimum. I can still walk through that code with the human and we can figure out what went wrong. I can ask the human about the code and I can tell a lot by their explanation, even if they make mistakes[2]. I can't trust the AI to tell an accurate account of even its own code because it doesn't actually understand. Even the dumb human has a much larger context window. They can see all the code. They can actually talk to me and try to figure out the intent. They will challenge me if I'm wrong! And for the love of god, I'm going to throw them out if they are just constantly showering me with praise and telling me how much of a genius I am. I don't want to work with someone where I feel like at any moment they're going to start trying to sell me a used car.

There's a lot of reasons, more than I list here. Do I still prompt LLMs and use them while I write code? Of course. Do I trust it to write code? Fuck no. I know it isn't trivial to see that middle ground if all you do is vibe code or hate writing code so much you just want to outsource it, but there's a lot of room here between having some assistant and having AI write code. Like the OP suggests, someone has got to write that 10-20%. That doesn't mean I've saved 80% of my time, I maybe saved 20%. Pareto is a bitch.

[0] Ever hear of "code rot?"

[1] Well... I'd rightfully dock points if they wrote obfuscated code...

[2] A critical skill of an expert in any subject is the ability to identify other experts. https://xkcd.com/451/

Re: How Anthropic teams use Claude Code

#124

Post by alwillis »

nextworddev wrote: post #26
lumost wrote: post #15

Earlier quoted context omitted.

once upon a time - engineers often had to concern themselves with datacenter bills, cloud bills, and eventually SaaS bills. We'll probably have 5-10 years of being concerned about AI bills before the AI expense is trivial compared to the human time.

You will start seriously worrying about coding AI bills within 6 months

You will start seriously worrying about coding AI bills within 6 months

Nope.

More open models ship everyday and are 80% cheaper for similar and sometimes better performance, depending on the task.

You can use Qwen-3 Coder (a 480 billion parameter model with 35 billion active per forward pass (8 out of 160 experts)) for $0.302/M input tokens $0.302/M output tokens via openrouter.

Claude 4 Sonnet is $3/M input tokens and $15/M output tokens.

Several utilities will let you use Claude Code to use these models at will.

Re: How Anthropic teams use Claude Code

#125

Post by politelemon »

maxnevermind wrote: post #107
chubot wrote: post #6

I use Claude and like it, but this post has kind of a clunky and stilted style So I guess the blog team also uses Claude

Also started to suspect that, but I have a bigger problem with the content than styling: > "Instead of remembering complex Kubernetes commands, they ask Claude for the correct syntax, like "how to get all pods or deployment status," and receive the exact commands needed for their infrastructure work." Duh, you can ask LLM tech questions and stuff. What is the point of putting something like that on the tech blog of t…

I think this is meant to serve as a bit of an advert/marketing and bit of a signal to investors that look, we're doing things.

Re: How Anthropic teams use Claude Code

#126

Post by theshrike79 »

aeontech wrote: post #117
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.

I mean... this is a deterministic task, can just run it through autoformatter? Why ask AI to do indentation of all things?

This is why I have a standing order for all LLMs to run goimports on any file they've edited. It'll fix imports and minor nags without the LLM having to waste context on removing a line there or changing ' to " somewhere.

Even better if you use an LLM with Hook support, just have the hook run formatters on the file after each edit.

Re: How Anthropic teams use Claude Code

#127

Post by theshrike79 »

bavell wrote: post #110
maerch wrote: post #89

Earlier quoted context omitted.

> A repeated trend is that Claude Code only gets 70-80% of the way, which is fine and something I wish was emphasized more by people pushing agents. Recently, I realized that this applies not only to the first 70–80% of a project but sometimes also to the final 70-80%. I couldn’t make progress with Claude on a major refactoring from scratch, so I started implementing it myself. Once I had shaped the idea clearly enou…

I need to try this - started using Claude code a few days ago and have been struggling to get good implementations with some high-complexity refactors. It keeps over engineering and creating more problems than it solves. It's getting close though, and I think your approach would work very well for this scenario!

Claude has a tendency to reinvent the wheel heavily.

It'll create a massive bespoke class to do something that is already in the stdlib.

But if there's a pattern of already using stdlib functions, it can copy that easily.

Re: How Anthropic teams use Claude Code

#128

Post by theshrike79 »

quintu5 wrote: post #56
jonstewart wrote: post #22

Earlier quoted context omitted.

The hilarious part I’ve found is that when it runs into the least bit of trouble with a step on one of its plans, it will say it has been “Deferred” and then make up an excuse for why that’s acceptable. It is sometimes acceptable for humans to use judgment and defer work; the machine doesn’t have judgment so it is not acceptable for it to do so.

My favorite is when you ask Claude to implement two requirements and it implements the first, gets confused by the the second, removes the implementation for the first to “focus” on the second, and then finishes by having implemented nothing.

This is why you ask it to do one thing at a time.

Then clear the context and move on to the next task. Context pollution is real and can hurt you.

Re: How Anthropic teams use Claude Code

#129

Post by theshrike79 »

duped wrote: post #119
jasonthorsness wrote: post #12

Claude Code works well for lots of things; for example yesterday I asked it to switch weather APIs backing a weather site and it came very close to one-shotting the whole thing even though the APIs were quite different. I use it at home via the $20/m subscription and am piloting it at work via AWS Bedrock. When used with Bedrock APIs, at the end of every session it shows you the dollar amount spent which is a bit dis…

Meanwhile I ask it to write what I think are trivial functions and it gets them subtly wrong, but obvious in testing. I would be more suspicious if I were you.

Ask it to write tests first and then implement based on the tests, don't let it change the tests.

Re: How Anthropic teams use Claude Code

#130

Post by theshrike79 »

throwawayoldie wrote: post #47
philomath_mn wrote: post #33

Earlier quoted context omitted.

Why is that?

Because at some point, Anthropic needs to stop hemorrhaging money and actually make some.

And when Anthropic raises the prices enough, people will jump ship.

That's why you don't pay the yearly license for anything at this point in time. Pay monthly and evaluate before each bill if there's something better out already.

Post reply on HN