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

Issue: Claude Code is unusable for complex engineering tasks with Feb updates

github.com

Post reply on HN
Print view View original

611–620 of 829 posts

Re: Issue: Claude Code is unusable for complex engineering tasks with Feb updates

#611

Post by andai »

faitswulff wrote: post #513
Wowfunhappy wrote: post #509

Earlier quoted context omitted.

> Under the hood, by setting this header we avoid needing thinking summaries, which reduces latency. You can opt out of it with `showThinkingSummaries: true` in your settings.json (see [docs]( https://code.claude.com/docs/en/settings#available-settings )). Can I just see the actual thinking (not summarized) so that I can see the actual thinking without a latency cost? I do really need to see the thinking in some form…

Anthropic's position is that thinking tokens aren't actually faithful to the internal logic that the LLM is using, which may be one reason why they started to exclude them: https://www.anthropic.com/research/reasoning-models-dont-say...

What's the implication of this? That the model already decided on a solution, upon first seeing the problem, and the reasoning is post hoc rationalization?

But reasoning does improve performance on many tasks, and even weirder, the performance improves if reasoning tokens are replaced with placeholder tokens like "..."

I don't understand how LLMs actually work, I guess there's some internal state getting nudged with each cycle?

So the internal state converges on the right solution, even if the output tokens are meaningless placeholders?

Re: Issue: Claude Code is unusable for complex engineering tasks with Feb updates

#612

Post by blurbleblurble »

hansmayer wrote: post #607
bcherny wrote: post #209

Hey all, Boris from the Claude Code team here. I just responded on the issue, and cross-posting here for input. --- Hi, thanks for the detailed analysis. Before I keep going, I wanted to say I appreciate the depth of thinking & care that went into this. There's a lot here, I will try to break it down a bit. These are the two core things happening: > `redact-thinking-2026-02-12` This beta header hides thinking from th…

You guys realise you are about 3 months into another one of your CEOs announcements that AI would "write all code in 6 months", right? Based on the problems you are facing, would you say your CEO gave a realistic announcement this time around ?

idk seems accurate from where I'm sitting

Re: Issue: Claude Code is unusable for complex engineering tasks with Feb updates

#613

Post by boesboes »

dboreham wrote: post #474
p1necone wrote: post #425

Earlier quoted context omitted.

The "this test failure is preexisting so I'm going to ignore it" thing has been happening a lot for me lately, it's so annoying. Unless it makes a change and then immediately runs tests and it's obvious from the name/contents that the failing test is directly related to the change that was made it will ignore it and not try to fix.

That said I've worked with several humans who did/said the exact same thing.

But did they say that about tests they just added themselves too? Had claude try that on me a couple of times >_<

Re: Issue: Claude Code is unusable for complex engineering tasks with Feb updates

#614

Post by oriettaxx »

areoform wrote: post #349
bcherny wrote: post #209

Hey all, Boris from the Claude Code team here. I just responded on the issue, and cross-posting here for input. --- Hi, thanks for the detailed analysis. Before I keep going, I wanted to say I appreciate the depth of thinking & care that went into this. There's a lot here, I will try to break it down a bit. These are the two core things happening: > `redact-thinking-2026-02-12` This beta header hides thinking from th…

Hey Boris, thanks for the awesomeness that's Claude! You've genuinely changed the life of quite a few young people across the world. :) not sure if the team is aware of this, but Claude code (cc from here on) fails to install / initiate on Windows 10; precise version, Windows 10.0.19045 build 19045. It fails mid setup, and sometimes fails to throw up a log. It simply calls it quits and terminates. On MacOS, I use Cla…

why is this post down graded?

Re: Issue: Claude Code is unusable for complex engineering tasks with Feb updates

#615

Post by AugustoCAS »

bazhand wrote: post #438
richardjennings wrote: post #375

Earlier quoted context omitted.

You cannot control the effort setting sub-agents use and you also cannot use /effort max as a default (outside of using an alias).

export CLAUDE_CODE_EFFORT_LEVEL=max

Thank you!

Worth mentioning that setting this via effortLevel in .claude/settings.json does not work. https://github.com/anthropics/claude-code/issues/35904

Re: Issue: Claude Code is unusable for complex engineering tasks with Feb updates

#616

Post by ehnto »

skippyboxedhero wrote: post #170

I appreciate the work done here. Been having this feeling that things have got worse recently but didn't think it could be model related. The most frustrating aspect recently (I have learned and accepted that Claude produces bad code and probably always did, mea culpa) is the non-compliance. Claude is racing away doing its own thing, fixing things i didn't ask, saying the things it broke are nothing to do with it, et…

I am still on an old version of CC on one machine, but the results are the same. More difficulty keeping it on track, convincing it timelines I suggest are correct etc. For example I had a deploy fail, and it would not believe that the new logs were not from a previous deploy. It was adamant it had fixed the issue, so the logs must be old logs.

Re: Issue: Claude Code is unusable for complex engineering tasks with Feb updates

#617

Post by OliverGuy »

johndough wrote: post #331
bcherny wrote: post #209

Hey all, Boris from the Claude Code team here. I just responded on the issue, and cross-posting here for input. --- Hi, thanks for the detailed analysis. Before I keep going, I wanted to say I appreciate the depth of thinking & care that went into this. There's a lot here, I will try to break it down a bit. These are the two core things happening: > `redact-thinking-2026-02-12` This beta header hides thinking from th…

I think it is hilarious that there are four different ways to set settings (settings.json config file, environment variable, slash commands and magical chat keywords). That kind of consistency has also been my own experience with LLMs.

settings.json -> global config Env vars -> settings different to your global for a specific project Slash commands / chat keywords -> need to change a setting mid chat

Re: Issue: Claude Code is unusable for complex engineering tasks with Feb updates

#618

Post by ramon156 »

Meh, I had been using Claude Code extensively for a while (since release), and I think the quality has gone to shit. I have no data to back up this claim, so it might be placebo.

GLM 5.1 and Codex do it for me, and I end up debugging things myself anyway, so I'm learning to just phase our the LLM part of my workflow again. Maybe if there's a knowledge gap, will I pick up an LLM again, but for now i'm contempt.

Re: Issue: Claude Code is unusable for complex engineering tasks with Feb updates

#619

Post by not_that_d »

andai wrote: post #611
faitswulff wrote: post #513

Earlier quoted context omitted.

Anthropic's position is that thinking tokens aren't actually faithful to the internal logic that the LLM is using, which may be one reason why they started to exclude them: https://www.anthropic.com/research/reasoning-models-dont-say...

What's the implication of this? That the model already decided on a solution, upon first seeing the problem, and the reasoning is post hoc rationalization? But reasoning does improve performance on many tasks, and even weirder, the performance improves if reasoning tokens are replaced with placeholder tokens like "..." I don't understand how LLMs actually work, I guess there's some internal state getting nudged with…

> I don't understand how LLMs actually work...

Plot twist, they don't either. They just throw more hardware and try things up until something sticks.

Re: Issue: Claude Code is unusable for complex engineering tasks with Feb updates

#620

Post by stingraycharles »

sothatsit wrote: post #589
stingraycharles wrote: post #577

Earlier quoted context omitted.

And I wonder how redacting them reduces latency, as it sure as hell doesn’t make the responses any faster and bandwidth isn’t the issue here.

They provide thinking summaries, so I assume they have to call Haiku or some other model to summarise the thinking blocks.

That’s not asynchronous? Wouldn’t it make more sense to disable those thinking summaries in those cases rather than hiding the thinking altogether?
Post reply on HN