@samlambert Claude design, with our organization design system. Just paste your thoughts, and let the magic operate. I always hated writing slides, now I must admit I like it.
@trq212: we are facing weird and random `Context Limit Reached` with the latest Claude Code version, even if we use only 13% of the context. Is it a known issue? Feedback: bf3ab2d0-6450-4f11-8197-22525c29200f
Thanks for this amazing version. Workflows are awesome! 🚀
I switched our code review skill (launching 6 agents in parallel) to the new #Opus workflow feature.
The gains are impressive: execution time cut in half, token usage reduced by 80%, and a higher reliability thanks to the JS execution.
Big shoutout to the @ClaudeDevs! 🙌
@GergelyOrosz "Hope this email finds you well" is part of my email filters. I don't want to spend time reading very generic AI slop.
I'm not against using AI to help redact documents, but it should be reviewed by the author. And never a human would let this unnatural sentence.
I am now able to launch #Claude within an isolated and secured @Hetzner_Online#VPS. It now runs in YOLO mode, without worrying about data exfiltration. As a bonus: I can now launch my own environment from my phone too! 🚀
https://t.co/7gwSBhIqab
@GergelyOrosz The only thing that made me keep a Windows was gaming. I completely switched a few months ago to Linux (CachyOS), and wow, gaming performances are even better than Windows... Not talking about a finally responsive OS. It's night and day!
@marcospereeira It does look to me like it is a benefit. If I'm a claude code user, you're telling me the apps and agents that have been piggybacking on a super cheap api that they get to slam for hella cheap murdering the api perf for me are going away? And I get a $200 monthly api cred? Cool
@benjamincode Idem. Si tu respectes leurs ToS, je pense pas que ce soit un problème. Ma compréhension est qu'ils les appliquent de plus en plus, pour éviter l'abus des plans Max (à la OpenClaw). Quitte à me faire backlash, je trouve ça fair pour garantir des perfs correctes pour tout le monde.
@GergelyOrosz As a non-native-English speaker, I often prompt "improve my English" to make my ideas clearer. I'm expressing my thoughts in a raw English and let #AI do its magic. That's a valid use case, as soon as the content is a real idea, not empty AI slop.
@txase@mikejulian My main concern with this approach is context bloat. I don't want the agent to have the 3-month history of all the changes provided to the "foo" feature. Just focus on the current implementation.
Didn't you face this problem storing everything in your repo?
@thsottiaux I played with the CLI an hour this morning for the first time (I'm a regular Opus user). My first two main issues are the quality of the Web Searches and the impossibility to scope a MCP to a given project.
¿Como estar actualizado en AI?
Para mi, la comunidad tech acá es de lo mejor para esto
Si quieren estar dos semanas adelante que linkedin, recomiendo a:
@trq212@bcherny - Claude code
@thsottiaux - OpenAI
@LLMJunky - Crack total en Codex
@karpathy - No necesita intro
Ustedes, a quien recomiendan?
At any other company, the person who just leaked 512,000 lines of source code, internal model codenames, and an entire product roadmap is clearing their desk by lunchtime.
Boris built Claude Code from a side project into $2.5B+ in annual revenue and 4% of all GitHub commits. His public response to the leak: the system failed, not the person. A manual deploy step should have been automated. One missing exclusion rule in a package config let a .map file slip into a routine npm publish.
The companies that fire the engineer who made the mistake get exactly one outcome: every other engineer learns to hide their mistakes. Near-misses go unreported. The next config error gets buried instead of flagged. You trade one public incident for dozens of silent ones.
Google codified this into SRE doctrine fifteen years ago. Every major outage postmortem at Google, Amazon, and Microsoft runs blameless by default because the math on incident frequency proves it works. The teams that punish errors have higher error rates. The teams that instrument errors have lower ones.
Anthropic just exposed their entire product roadmap to every competitor in the space. The response from the head of the product: here's what broke in the process, here's what we're automating, here's what ships next to prevent it.
That's how you keep the engineers who build billion-dollar products from leaving for the company that won't fire them over a config file.
@gameromandev@AnthropicAI@claudeai I honestly don't care. It's most likely an error, and I don't want to waste time to counter it, as it seems like a lot of incoming bureaucracy: https://t.co/SS5unJ1B35
I can live without this fork.
I forked this public official repository from @AnthropicAI: https://t.co/ArxTaUomj2. I wanted to see if I could help with some @claudeai issues, as it seemed open-source.
Today, I'm receiving a #DMCA notice? It looks super aggressive for an open-source repo, isn't it?
@ai_sentience@trq212@joelhooks That's most likely a mistake, but receiving this kind of threatening messages ("we suggest you consult a lawyer") is never a pleasant experience.
@AnthropicAI@claudeai It's most likely related to the fork that appeared yesterday. But this repository has nothing to do with it. The fork has been created weeks ago. I guess it's a mistake?
@claudeai is becoming unusable. I’ve hit 54% of my limit in less than an hour, despite never hitting the $200 plan ceiling before the update.
This is a risky move by @AnthropicAI. Many engineers are already exploring alternatives, including local models.