John

Woah - this is fantastic @figma designsystems.com/open-design-sy…

Claude Design @claudeai is 🔥 Take the time to set up the design (brand) system first, then let the vibe designing fly.

I can't go back to the regular YouTube UI after this 😅 Obsidian Reader now makes the transcript interactive so you can scrub, highlight, auto-scroll. It feels so nice.

Trying `--dangerously-skip-permissions` for tightly scoped experimental parts of the app and its working well so far.

@amorriscode @bcherny .worktrees and localhost:300(n+1) with blow.aiff sound alerts when it needs my attention

if you're bored waiting for claude to finish doing work, start another session. life is too short to be bored.

no 👏 more 👏 permission prompts 👏

@Mappletons , as always, is so spot on maggieappleton.com/now-2026-01 . It's the most wild shake up of design, engineering, SWE ... and for those of us with young ones at home, even wilder of a life change. Keep writing Maggie!

Anyone know who’s behind @AnthropicAI’s solutions like this? claude.com/solutions/fina… . What they did here is really smart

@ivanhzhao always with the spot-on technical & industrial design metaphors notion.com/blog/steam-ste…

@ivanhzhao always coming with the comprehensive technology metaphors notion.com/blog/steam-ste…

#progress monospace terminal art by @claudeai Code and @vercel

@pablostanley @v0 Having a mid-stack portal to invite design-minded collaborators (in-house, clients, agencies, etc.) into your codebase is HUGE. Super excited to play more with this. Question - is there a way to put caps on token usage by user/group?

in case you didn't know... in @v0 you can connect to any github repo and branch, and work on it directly

There’s a lot of noise from the AI hype crew, as well as the doomers. I’m more interested in the quiet middle. There’s a group of builders who are too busy building, grappling with the meaning of work, the societal impact, and helping close friends. I am building.

AI SDK 5 Introducing type-safe chat, agentic loop controls, data parts, speech generation and transcription, Zod 4 support, global provider, and raw request access.

@thecodaguy @figma @SubframeApp @cursor_ai @AnthropicAI @claude_code @nextjs @tailwindcss @ThePSF @supabase @neondatabase @vercel @Railway For the LLM? Use case (sometimes cost/quality). Claude (Opus 4) for code (Sonnet) for analysis and writing, ChatGPT for chatting & research, Gemini for huge context windows, Perplexity for search. You?

@JohnScrugham @figma @SubframeApp @cursor_ai @AnthropicAI @claude_code @nextjs @tailwindcss @ThePSF @supabase @neondatabase @vercel @Railway What does it depend on???

My stack is: - @figma : Slides, brand, mocks - @SubframeApp: frontend & design systems -> code - @cursor_ai (b/t their agent and @AnthropicAI 's @claude_code ) w/ @nextjs , @tailwindcss , @ThePSF - @supabase or @neondatabase - Deploy: @vercel or @Railway - LLM: it depends

Why city flags may be the worst-designed thing you've never noticed: t.ted.com/DTSUmLm @romanmars

Google Meet live presentation feedback

Building an #AI “wrapper” SaaS #app, what #nocode tool has the best @OpenAI connection?

@coda_hq I have built (over time) a hub for my business and have been using it with clients to help organize their businesses. It’s like a one-stop shop for all things that you need to think about or do in your business. It’s a really long list. : )

The Strangely Beautiful Experience of Google Reviews longreads.com/2023/01/03/the…