So every time I say the word 'workflow' in Claude Code...
(let's say, when I'm creating a new GitHub workflow)
...it tries to enter 'workflow' mode, spinning up dozens of subagents to complete my task.
Stupid fucking thing
It's incredibly annoying that Claude decides to launch multiple sub-agents on its own just for mentioning the word "workflows" in the prompt.
Not /workflows, no, just mentioning it. Even in a context where it doesn't make sense for it to execute.
It's too common a word, and they're just parsing words in the prompt.
Please @ClaudeDevs remove that. It's very annoying because it doesn't warn you and it consumes all the tokens.
To make matters worse, there isn't even clear feedback in the Claude Code VS Code extension like there is in the Claude Code CLI.
Introducing Claude Opus 4.8: it builds on Opus 4.7 with sharper judgment, more honesty about its own progress, and the ability to work independently for longer than its predecessors.
Available today at the same price.
Los que nos fuimos en 2019-2023 pensamos que mejoraríamos la calidad de vida de nuestra familia en Cuba y terminamos pagando para intentar que vivan casi cómo estaban cuando nos fuimos.
✨Socialismo✨.
What he says is not necessarily like that and has nuances.
The macbook are infinitely better than almost any laptop window, and proof of this is the same neo that having 8 GB RAM and a phone chip performs better than many windows with more resources.
Composition, quality, hardware, software, stability and integration are far above.
The professional sector mostly prefer to use mac and to find a window with the same quality you usually pay the same or more. Now, that most non-professional people buy a Mac just because it's Apple is something else. But that comparison of 1800 usd against 300 is an aberration of comparison
One of the most influential and respected people in the AI world has joined Anthropic.
Andrej Karpathy, one of the early founders of OpenAI and ex Director of AI at Tesla.
An absolute machine.
Personal update: I've joined Anthropic. I think the next few years at the frontier of LLMs will be especially formative. I am very excited to join the team here and get back to R&D. I remain deeply passionate about education and plan to resume my work on it in time.
@DanielBlancoSWE yo hago un cambio grande mientras hago otros pero mas pequeñitos que no requieran mucho esfuerzo mental para mi.
Tambien te digo, yo antes programaba el fin de semana pero ahora, cuando acaba la semana ni ganas de abrir el pc tengo
a few months ago, the standard for agents was using 200k tokens in context window. Now it's 1 million, and it's practically mandatory when you're using coding agents.
today I tired Cursor again on a Pro account capped at 200k tokens, and the context disappears fast. Just one request to fix a few things in the code ate up 60% of the context
Big news: I’ve been selected as @cursor_ai Regional Lead for Mexico & Central America
Calling all Mexican AI builders!
I’ve got some credits if you wanna build with Cursor!
Let’s build 🚀
Multiple security vulnerabilities affecting React Server Components and Next.js have been disclosed. We strongly recommend updating your applications immediately.
Cloudflare WAF managed rules already mitigate the disclosed denial-of-service vulnerabilities, and we are investigating additional coverage for several other CVEs.
https://t.co/mT9ujk1H7c
Introducing GPT-Realtime-2 in the API: our most intelligent voice model yet, bringing GPT-5-class reasoning to voice agents.
Voice agents are now real-time collaborators that can listen, reason, and solve complex problems as conversations unfold.
Now available in the API alongside streaming models GPT-Realtime-Translate and GPT-Realtime-Whisper — a new set of audio capabilities for the next generation of voice interfaces.