galbc

@TheAmolAvasare @justo_grane @Elemont @patomolina Same here. We have not heard back from Anthropic

@justo_grane @Elemont @patomolina You’re welcome 🙏

@patomolina Looking into this

@opencode Scratch that - we got it this morning

@opencode Hi! It's not appearing for our team

Kimi K2.6 now in OpenCode — Go included

@jonoringer I've been quite impressed by kimi-k2.6 thus far. Eager to push its limits

kimi-k2.6 in hermes is solid. Chinese open source models are moving fast.. where is Llama/Meta?!

@TheAmolAvasare @justo_grane @Elemont @patomolina Our organization is in the same boat. 40+ org banned without warning

@justo_grane @Elemont @patomolina Yes will take a look

@Elemont @TheAmolAvasare @patomolina This feels quite topical

@TheAmolAvasare @patomolina Could you reply to me about my ban please, or do I need 700k views on X first?

@TheAmolAvasare @patomolina The same thing happened to our 40+ team. We're a professional organization and have dedicated time to investing in Claude as our primary operations hub. We've received no support after being banned without warning.

@patomolina @claudeai They did the same to us today with no response @AnthropicAI

@claudeai you took down our entire organization with 60+ accounts belonging to a legitimate company for no apparent reason, without any explanations. The only way to appeal the decision is by filling out a Google Form? Very bad UX and customer service.

@trq212 @bcherny @claudeai @ClaudeDevs - Our organization's account was suspended without warning, disrupting our operations and product. We've submitted an appeal but this is an urgent matter. Is there anyone from the team that can help expedite review?

> Not really real ShinyHunters > Claims to have compromised Vercel > Real ShinyHunters say "wtf that's not me" > Impersonator ShinyHunters says stole source code, customer data, databases etc > Vercel makes security bulletin > Announces compromise > Real ShinyHunters "wtf that's not us tho fr" 1. WHO EXTORTS SOMEONE ON A SUNDAY 2. 200iq move to blame ShinyHunters for compromise 3. 400iq move if ShinyHunters made fork of ShinyHunters claiming to be impersonator ShinyHunters to convince everyone the fake ShinyHunters are impersonating ShinyHunters, but it was actually ShinyHunters being the fake ShinyHunters all along 4. Lots of cybercrime drama right now, but ITS SUNDAY. Dawg, WAIT UNTIL LIKE TUESDAY OR SOMETHING. Smdh

@marclou Enjoy the rest! Good look with hyrox. My brother just did his first and said the wall balls were the toughest movement.

I overworked myself. For the past 5 years, I barely took a day off. I shipped 30 startups, wrote 30,000 tweets, made 70 YouTube videos, and answered thousands of support requests. The weird part is, it never felt like work. It felt like play. Until recently. For the first time in years, I’ve struggled to get things done. I don’t want to open my laptop. The things that used to make me happy suddenly feel heavy. That scares me a bit, because the playful side of work is what got me here. And right now, it feels like I lost it. So I’m taking a break. I’ll still do about an hour of maintenance work a day, but the rest of my time will go to real life. I’m training for Hyrox in Korea on May 25, so I’ll put my energy there for now and train 20 hours a week. I went through a little burnout once before, in 2021. What helped me was going back to basics: training, reading, eating well, and sleeping 8 hours. It brought the hunger back then. I trust it will again!

@anothercohen My recommendation would be The Top, then Kin for a cocktail. Other honorable mentions: Swamp if you want nostalgia. Cry Baby's Indian Street Food if you want Dosa Dragonfly for sushi Germain's for a chicken sandwich

Where should I go to dinner in Gainesville, FL?

@jennyrozelle "What's luv? It should be about us, it should be about trust"

Insider tip (from an estate attorney): If you set up a Trust, you can name it whatever you want. Sure, you can go with something standard like “Smith Family Trust”… but you don’t have to.

@sandislonjsak If you plan your projects well, you can review much faster because you have an expected outcome. Deviations and regressions become blatantly obvious.

@nikitabier @benjitaylor The most important action you can take to improve this application is to verify each account is a real human being or business entity. End the bots. End the slop.

Ladies and gentlemen, I'd like to welcome @benjitaylor to 𝕏, our new design lead. I met Benji six years ago when I invested in his app: it was one of the most well-designed products I'd encountered. I knew right away he was on track to become one of the best designers in the world. After 6 months of convincing, we're finally teaming up and building the greatest design team in the industry.

@JohnThilen @doodlestein It's a choice, of course. However, to avoid disruption, spinning up a worktree to debug or review a PR is an efficient way to switch contexts. It works well for our team, but at the end of the day, different strokes! I use worktrees to build across many projects at once.

@galbc1 @doodlestein To be precise: it is possible that some of them use git worktree to organize their agents. I don't want to mess with what works for them. The point is that they don't use git worktree to collaborate with other humans.

It's fascinating to me how different my workflows are for using Claude Code (and Codex) compared to most of the ones given here. I categorically reject their biggest tip, for instance, and attribute to that much of my productivity using these tools. Lesson: think for yourself!

@doodlestein This is likely why you're not resonating with their recommendations. When working with a team of humans, each of which may have a team of agents, the paradigm changes. Worktrees are absolutely necessary for quick code reviews, parallel work, etc.

@galbc1 A team… of robots.

@bcherny @charles_irl Does the same apply for skills?

In case it’s not clear in the docs: - Ancestor CLAUDE.md’s are loaded into context automatically on startup - Descendent CLAUDE.md’s are loaded *lazily* only when Claude reads/writes files in a folder the CLAUDE.md is in. Think of it as a special kind of skill. We designed it this way for monorepos and other big repos, tends to work pretty well in practice.

this is a crazy policy to me -- loading every CLAUDE md up the tree AND down it

@PovilasKorop Spend more time planning. Then, the review is more about glaring deviations from the plan. Furthermore, with proper configuration and agentic review, those deviations happen less and less.

With all AI hype, I still don't understand how people run multiple Claude Code agents/terminals. Ok, I understand how it's possible technically. But how do you REVIEW all of that? I don't have enough brain capacity for such multi-context-switching. Or am I just too old? :)