Bryan Housel

maintainer of @mapwithrapid and other open source mapping curios

[Bryan Housel @bhousel]

🧚‍♂️ Pixi v8 Upgrade!
We upgraded Rapid’s render to use the latest version of @pixijs game engine. This version gives Rapid a welcome performance boost - if you’ve aired any grievances about Rapid being laggy, the new Rapid should perform much better for you!
2/6

[Bryan Housel @bhousel]

LMAO @GitHubCopilot this is ridiculous.

The thing you are offering in June isn't anywhere close to the thing I signed up for in January.

[Bryan Housel @bhousel]

@DirtyTesLa My car did this once but I’m pretty sure it was just reacting to other cars on the highway doing the same thing? It would be like if everyone was changing lanes to avoid a debris, FSD will do it too.

[Bryan Housel @bhousel]

@drtheglide @DirtyTesLa I've just been leaving them foul-mouthed voice memos about how much I hate the screen.

[Bryan Housel @bhousel]

@SawyerMerritt I hate this screen so much, I just send them angry voice memos now whenever it appears. I don’t need to fill out a feedback form every time I get home.

[Bryan Housel @bhousel]

Npm pushing trusted publishing / OIDC was a mistake. So glad I never opted any of my projects into it. I will probably forever be manually publishing projects using short term 2FA tokens, and I feel good about that.

[Adnan Khan @adnanthekhan]

This attack leveraged GitHub Actions Cache Poisoning.

Payload deployed here: https://t.co/TawmjgwdBB

It looks like it detonated here: https://t.co/Cfeoa4aSgo

[Todd Saunders @toddsaunders]

I want to get the most AI pilled folks in NJ for a meetup in Westfield.

I have no idea how to source or find these people so please X do your thing!

[Bryan Housel @bhousel]

@toddsaunders I’m in! DMs open

[Feross @feross]

🚨 CRITICAL: Active supply chain attack on axios -- one of npm's most depended-on packages.

The latest [email protected] now pulls in [email protected], a package that did not exist before today. This is a live compromise.

This is textbook supply chain installer malware. axios has 100M+ weekly downloads. Every npm install pulling the latest version is potentially compromised right now.

Socket AI analysis confirms this is malware. plain-crypto-js is an obfuscated dropper/loader that:

• Deobfuscates embedded payloads and operational strings at runtime
• Dynamically loads fs, os, and execSync to evade static analysis
• Executes decoded shell commands
• Stages and copies payload files into OS temp and Windows ProgramData directories
• Deletes and renames artifacts post-execution to destroy forensic evidence

If you use axios, pin your version immediately and audit your lockfiles. Do not upgrade.

[Bryan Housel @bhousel]

I asked my daughter about this, she uses I-Ready for school. She showed me that there are some cutscenes that you can’t skip, but when it starts speaking, you can click the speaker icon to skip the narration. She doesn’t like it but she has learned how to speed run through the lessons.

[Bo Wang @BoWang87]

Prof. Donald Knuth opened his new paper with "Shock! Shock!"

Claude Opus 4.6 had just solved an open problem he'd been working on for weeks — a graph decomposition conjecture from The Art of Computer Programming.

He named the paper "Claude's Cycles."

31 explorations. ~1 hour. Knuth read the output, wrote the formal proof, and closed with: "It seems I'll have to revise my opinions about generative AI one of these days."

The man who wrote the bible of computer science just said that. In a paper named after an AI.

Paper: https://t.co/juSOmK9vOt

[Aakash Gupta @aakashgupta]

The buried lede in this post is the October-to-December timeline.

In October 2025, Karpathy publicly said AI agents “just don’t work.” Eight weeks later he’s 80% agent-coded and calling it the biggest workflow change in two decades. That’s the fastest opinion reversal from one of the most credible voices in AI, and it maps to something measurable.

Stack Overflow’s 2025 developer survey tells the other side of this story. Only 16% of developers reported “great” productivity gains from AI tools. 45% said debugging AI code takes longer than writing it themselves. Meanwhile, the Claude Code team is shipping 20-27 PRs per day, 100% AI-written.

That’s a bimodal distribution forming in real time. A small group is getting 10x leverage. The majority is getting modest autocomplete improvements and spending extra time fixing hallucinated code. Same tools, completely different outcomes.

Karpathy names the variable: “ascending the layers of abstraction to set up long-running orchestrator Claws with all of the right tools, memory and instructions.” That’s a skill that looks nothing like traditional programming. You’re managing agents the way a senior engineer manages junior devs. Scoping work, reviewing output, catching failure modes, maintaining system-level context.

The 10x engineer ratio he’s worried about is already here. 25% of YC’s Winter 2025 batch shipped codebases that were 95% AI-generated, and every founder was fully technical. They weren’t replacing skill with AI. They were compounding skill through AI. The gap between someone who can decompose a weekend project into agent-executable chunks and someone still typing code line by line is widening by the month.

What makes this post different from the usual AI hype: Karpathy is explicitly naming the failure modes. Needs high-level direction. Needs taste. Needs judgment. Works better for well-specified tasks. This is an honest field report from someone who mass-reversed his own position 90 days ago, and that combination of enthusiasm and specificity is what makes the signal real.

[Bryan Housel @bhousel]

I, too, ruminate on all my past dumb mistakes. My agent is just like me, but with a markdown file.

[Corey Ganim @coreyganim]

The secret file he didn't mention:

`. learnings/LEARNINGS. md`

Every time my agent makes a mistake, it logs the correction and updates its own rules.

43 skills. 661 lines of learnings. An agent that gets smarter every day.

Your setup isn't just a moat. It's a flywheel.

[Bryan Housel @bhousel]

Claude actually - probed into VSCode's internal chat storage, figured out the storage format, and wrote some temporary python programs to recover the missing chat. I'm still in shock 😳🤯 WTF! Never seen anything like this.

[Bryan Housel @bhousel]

Well this is a new one. I was working on a pretty extensive plan in @code and clicked "Continue in Background". The plan just disappeared from the chat as if it never happened. But wait..

[Bryan Housel @bhousel]

Then, I just asked it "I haven't seen VSCode lose a substantial chat before. Try to think of ways that we could get it back"... 🤨

[Bryan Housel @bhousel]

@kenwheeler i’ve been trying to rebrand it as “the secret language of the universe and ultimate truth” but my kids aren’t buying it.

[Bryan Housel @bhousel]

This will end OpenAI, if they really do decide to go this route.

[The Information @theinformation]

OpenAI’s CFO says the company will take a cut of the IP its AI helps customers invent.

[Ben Krasnow @BenKrasnow]

I recently learned the USGS has free aerial LiDAR data for many areas. Draw a box, then click the green button on the right side and select Eptium viewer. https://t.co/iY4QCNQ12S