Gent B

I made an AI graveyard for when people hit their Claude Code limits aicemetery.xyz #claudecode

the 90x multiplier is real but the bottleneck shifts from writing code to reviewing and understanding what the AI wrote. the devs who win with this aren't the ones generating the most code — they're the ones who can read it critically and catch the subtle bugs AI introduces confidently

Want to make new software the way I do? I only started 72 days ago. I am now on pace to do 90X the amount of software engineering than I did the last time I was working hard on code in 2013. Literally with AI it's like I have 90 versions of myself coding.

@0xdoug it's checking stdin/tty state to decide if it's running interactively or piped into something. focused terminal = interactive mode with richer output. unfocused or backgrounded = it assumes non-interactive and strips back. smart architecture choice tbh

It’s interesting that Claude Code seems to behave differently depending on if your terminal window is focused or unfocused

@icanvardar benchmarks measure isolated tasks but claude code's real strength is multi-file context and long session reasoning. the harness doesn't test what actually makes it good in practice

how is claude code the worst harness in opus 4.6 benchmarks 💀

@zachlloydai Codex is good at code quality checks

@_GentB haha totally...What's working for you lately in our code base to not write slop?

anthropic just accidentally shipped their entire claude code source code to npm. 512,000 lines. 41,500 github forks before the DMCA notices hit. i use this tool every day. here's what the leak actually tells us: it's not a chatbot. it's a multi-agent system with persistent memory, self-review capabilities, and anti-distillation DRM baked in. the leak doesn't matter. the moat was never the code. it's the model + the ecosystem of 100K developers who already built their workflows around it. $2.5B in annual revenue. from a terminal app most people don't know exists. that's the real story.

@AlexPalcuie at least the snail sticks around when you hit usage limits

help, i typed /buddy in claude code and now a snarky snail judges me through every outage

@trq212 they really added a gacha system to a CLI tool before fixing usage limits

/buddy

@chifeng1113 "selling maps in the age of self-driving cars" is an insane line lol. i switched from cursor to claude code a few months ago and haven't looked back. the difference is cursor helps you write code, claude code just... writes the code.

Cursor is in serious trouble. They built a $30B business on assisted coding. Anthropic's Claude Code just made that model obsolete. The market doesn't want a copilot anymore. It wants a pilot. Cursor is selling maps in the age of self-driving cars.

The dirty secret of AI coding in 2026: nobody can tell the good code from the slop anymore. Not maintainers. Not reviewers. Not even the engineers who shipped it. We have a quality crisis and the industry is pretending it isn't happening.

Someone said this was life before Anthropic nerfed Claude usage limits

@IntCyberDigest technically npm already did this for them yesterday

🚨‼️ BREAKING: Anthropic has decided to open source their entire codebase and is rebranding their AI to OpenClaude. Anthropic CEO Dario Amodei said: "Yesterday was no slip-up. If we disappear just like OpenAI is vanishing right now, our code can live on through the community."

@AlexPalcuie the accidental open source gave us more insight into how they build than any blog post ever did. respect for owning it like this

I repeat this to every new joiner at Anthropic but it's worth repeating in public -- we have a blameless culture and no single individual is at fault when bespoke complex systems break at scale

@Polymarket investors finally reading the source code that leaked and realizing Anthropic is actually shipping

JUST IN: OpenAI shares have reportedly “fallen out of favor” on the secondary market, as investors are pivoting to Anthropic.

@DeryaTR_ we're all just beta testers bouncing between Claude and Codex seeing who runs out of tokens first

Presumably Anthropic had fixed the Claude Code bug, so I wanted to give them the benefit of the doubt one more time & compare its rate limits with Codex. I used this simple prompt for planning and building: “Build a cute pixel RPG to play on the iPhone.” Both my Claude Max account and OpenAI Pro account started with fresh 5 hour resets. After several follow-up prompts, Codex completed a decent, simple, but fully playable RPG with exploration, crafting, collecting, fighting enemies & NPC interactivity features for the iPhone. I still had 93% of my 5 hour rate limit remaining with Codex using 5.4 xhigh & had used only 2% of my weekly limit. Claude Code, by contrast, still had not produced even a minimally viable playable version, with most of the above features missing, after already consuming 80% of the 5 hour limit. I asked for one more revision, but ran out of the 5 hr limit before it could even complete that. I then waited for a new reset and tried again to see whether it could finish the minimal game. It used up another 12% of the 5 hr limit and claimed it had added all the missing features. Yet the game was exactly the same, still unplayable, and not a single one of those missing elements had been added. Claude had also already used 8% of my weekly limit, still without delivering a product that Codex had already completed. Given that Anthropic also did not even bother to reset the limits from the token eating bug, I have now decided to cancel my Claude Max subscription. It has been a very frustrating waste of time dealing with it, and it has consumed all of my goodwill toward Anthropic.

@Rahatcodes my is_negative score is probably off the charts after hitting usage limits 4 times today

Claude Code has a regex that detects "wtf", "ffs", "piece of shit", "fuck you", "this sucks" etc. It doesn't change behavior...it just silently logs is_negative: true to analytics. Anthropic is tracking how often you rage at your AI Do with this information what you will

@kevinxu no thanks im not exit liquidity

Would you rather: - Go all in OpenAI's IPO - Go all in Anthropic IPO

@MeekMill Meek you using Claude Code?

When I start doing a lot for my community I start experiencing influenced extreme hatred. When I was just wilding and doing me just making money it was all good!

@DataJunkie great approach

Having to go AI first in my class (by choice) allows us to spend more time talking about systems (more interesting and more my expertise) rather than syntax.

@jacob_posel @bcherny @claudeai They're working on it. but I do wish they would 2x usage again or at least reset limits while it's an issue :/ I built a graveyard for people who hit their Claude Code limits aicemetery.xyz

Hey @bcherny @claudeai I'm on the $200/mo plan and blowing through usage instantly. Doesn't feel right. Is there any way to audit my account? Unfortunately I have experienced several bugs with the Claude product and I fear my plan configuration is not correct. Thanks

@adxtyahq TIME FOR ANOTHER MAX ACCOUNT! bury your prompt at aicemetery.xyz

ran claude-opus-4.6 high-thinking on a small prompt 760k tokens, 11m 27s runtime cost me ~$12 for a single call 💀

So everyone's hitting usage limits on Codex too now???? Must be the reach Claude Code usage limit -> hop on Codex pipeline BUT THEY'RE GETTING THEM RESET!?