Julian

@theo I can see switching to obsidian cleared your head

Just posted an important announcement on my LinkedIn profile

@theo Happy birthday!

This is the greatest birthday present of all time

AI-maxxing at its finest

@Baconbrix I'd love to play around with this! My UN is buishi thanks for making cool stuff!

Expo Agent just let me one-shot a custom lock screen widget Literally never been easier to create any native feature

@wp_acf When does ACF 6.8 come out of beta?

🚀 ACF & ACF PRO 6.7 now available. ✨ Say hello to Inline Editing for ACF Blocks - our most intuitive block-building upgrade yet. Content editors can now edit ACF Block fields directly in the preview, no side panel, no switching, with zero additional JS or custom markup required. 👇 Link in the reply.

@theo all of the nano banana models seem to be hanging on T3 chat, thought you should know.

@DavidKPiano @EffectTS_ It’s this kind of beauty

TypeScript devs will say stuff like "have you ever seen more beautiful code" and it literally looks like this

This is the future: AI that responds not just to explicit instructions, but also human intent and emotion. LLMs will take a big step when they remember and learn from emotional intent (e.g. user is angry: won’t do that again)

@theo oh geez. this is bad.

Top 5 apps right now: - 3 AI chats - TEMU - “Free cash” Idk how to explain but this feels like a recession indicator

@poetengineer__ this is so cool!

all my obsidian notes are now a living, digital garden 🌿🌸 each plant is a notes from a tag: older ones on the trunk, newer ones as leaves. i wanted to create a sense of tending your garden, so scrubbing the timeline lets you watch your notes grow chronologically.

@theo Also, Playwright is too slow for closed-loop e2e testing of UI.

@theo I feel like verification is the hardest problem right now preventing from reaching the "next level". A closed loop dev flow around react native is difficult, for example, because I'm unaware of good tools for closing the loop. Though, I'm thinking about looking at RNTL again...

I would like to purchase a handful of code problems that modern LLMs can’t solve. Requirements: - programmatically verifiable (can be tested without human interaction) - “before” state (repo before the commit that implements the solution) - example code that actually solves the problem I am willing to pay up to $500 per problem that I can easily test locally and confirm current models (gpt-5.3-codex, opus 4.6) are unable to solve. If you can’t tell, I’m running out of “too hard for LLM” code tasks 🙃🙃🙃

@benjaminbardou Are your artwork viewable in 3-D, i.e. with a VR headset? Or are they generated videos?

I dream of a piece that would be nothing other than a trajectory: a path moving between points of light, like stars scattered across a space of memory. The lines connecting these luminous centers would draw a shifting constellation. The figure born from these tensions would become the work itself — not an isolated image, but the very trace of its appearance, the vibrating cartography of a flow. [The Flow / work in progress]

Apparently Anthropic is relaxing their safety requirements. Per their article "Despite rapid advances in AI capabilities over the past three years, government action on AI safety has moved slowly." Wonder why?

@abiosoft Releasing into the public domain: github.com/JulianKingman/…

I got fed up with Docker Desktop, so I made a lightweight UI for Colima by @abiosoft Thanks Abiola, my docker setup feels much lighter now!

@DavidKPiano Thank you, I was confused

(this is sarcasm btw)

I don't like how everyone is punching down on Anthropic for this From max API keys being used too "maximally" to words that sound like "Claude" and now data that obviously only belonged to them, they're clearly struggling from theft and I feel bad for them

Oh dear anthropic, I don’t think this is going to work out the way you think it is. To say you’re unlikely to get sympathy for what looks like copyright abuse would be an understatement.

@theo F-ing dork here, he’s above the target 🎯

This is the top. Only downhill from here

@hasantoxr do you have a source for this?

🚨 The creator of Claude Code just leaked how his own team actually uses it. Boris Cherny (Anthropic engineer who BUILT Claude Code) posted his internal workflows on X. Someone turned it into a CLAUDE .md file you can drop into any project right now. No guessing how to use Claude Code anymore. Here's what's inside: ↓ → Workflow orchestration (how Boris's team structures complex tasks) → Subagent strategy (how they split work across multiple Claude instances) → Self-improvement loop (Claude learns from YOUR corrections over time) → Verification before done (no more half-finished outputs) → Autonomous bug fixing (it finds and fixes its own mistakes) → Core principles straight from the team that built it The compounding effect is wild. Every correction you make gets captured as a rule. Every rule drops Claude's mistake rate. Every project gets better than the last. This isn't a prompt template. It's a system that gets smarter the more you use it. Built by the guy who created the tool. Shared publicly on X. Packaged into one file you can use today. If you ship with AI daily, this CLAUDE .md will save you more time than any other file in your codebase.