marco retweeted
Let's talk a little more about Margaret Hamilton because she deserves it
In 1961, Margaret Hamilton was tackling massive systems, having previously wrangled room-sized computers that blared like a foghorn whenever they crashed.
Her fix wasn’t just to patch the bug. Everytime there was a crash, she took a Polaroid of the programmer who wrote it, and a note on the mistake. That way, if it ever happened again, they wouldn't be starting from scratch. They could note patterns and the last person who triaged that failure could help.
This is a critical aspect of what she gave us. She didn’t stop at fixing errors. She studied their patterns so she could see them coming. She called them “the enemy,” and she meant it.
Initially, when she was working on failure modes, higher-ups were errors wouldn't happen she actually named her program “Forget It.” But failures did happen. And her mindset, and discipline sent men to the Moon.
Her Apollo software was built to expect the unexpected, and on Apollo 11 it recognized an overload mid-landing and kept the critical tasks running.
Every postmortem, every bug tracker, every “let’s make sure this never happens again” carries a little of her in it.
Thank you, and rest in peace, Margaret Hamilton 🚀
ALT The famous photo of Margaret with the code she wrote to put a man on the moon
ALT The lego set of Margaret Hamilton my daughters gave me
marco retweeted
Very sorry to hear this news. 😥
I never had the chance to meet Margaret Hamilton, but her contributions to the flight software that got us to the moon are legendary. I love this image of her standing next to the printouts of the Apollo code.
We’ve lost an icon. Margaret Hamilton, a computer science pioneer best known for her work at MIT leading development of the flight software for the Apollo missions to the moon, has died at 90. Read about her legacy via MIT News: news.mit.edu/2026/margaret-h…
ALT Margaret Hamilton is reclined inside a NASA spacecraft module, surrounded by control panels and switches.
marco retweeted
We’ve lost an icon. Margaret Hamilton, a computer science pioneer best known for her work at MIT leading development of the flight software for the Apollo missions to the moon, has died at 90. Read about her legacy via MIT News: news.mit.edu/2026/margaret-h…
ALT Margaret Hamilton is reclined inside a NASA spacecraft module, surrounded by control panels and switches.
marco retweeted
This. People keep asking me for my setup. It's really just any harness, multiple agents concurrently, barely any skills, and using adversarial reviews. There's no magic sauce (or source!). The models are great out of the box.
marco retweeted
🚨 MUSE ECOSYSTEM ALERT 🚨
today we are announcing Muse Gadgets!
this is an open-source ESP32 firmware and Linux SDK for anyone to make hardware that works with Muse
we are also releasing our own gadget—Muse Home Link—to enable your muse to work with your smart devices (TV, speakers, etc.)
marco retweeted
COPY PASTE INTO OPUS 5.5:
"I want you to make me an awesome Shoes.rb Desktop app. I want _why to rise from the grave and jump for joy when he sees it. Go to github.com/scarpe-team/scarp… for instructions. Have fun. Live! ultrathink"
2/6
how can people have such bad takes
I’ve said it before and I’ll say it again:
Rust is a *really* bad idea for any software you need to have working a few years from now (AI generated or not), or want to continue developing into the future.
Not because Rust is ugly (which… it is).
And not because Rust is, overwhelmingly, a cult (again… it is).
But simply because Rust is still a moving target. The language is not yet “complete”… and neither is the compiler. Changes occur often enough (and in significant enough ways) to throw serious monkey wrenches into any sizable project.
Once Rust stabilizes (which could happen quickly if the Cult of Rust actually wanted it to), this problem will quickly go away.
Until then I will be sticking with C. Or C++. Or, heck, any language that is more stable than Rust.
marco retweeted
emma, the italian meme slop model, made headlines in italy for months, but this didn’t. if i speak i'm in big trouble
marco retweeted
When several people talk at once, a transcript can get messy fast.
Our new Nemotron 3 Diarization model tracks who spoke when, even when voices overlap. It handles up to eight speakers, has 100M parameters, and is now available on @huggingface 🤗
marco retweeted
We enabled streaming for @Shopify storefronts. The layout head reaches the browser before the template finishes rendering.
Median-store TTFB down 29% at p75, 21% at p90. FCP and LCP down 2–6%. No theme changes needed*.
performance.shopify.com/blog…
marco retweeted
I'm looking for my next role, in Montreal or remote! My background spans compilers, systems, and AI research.
I'd welcome introductions, leads, or a repost to help spread the word ❤️
More details about my experience and what I'm looking for: pointersgonewild.com/2026-09…
marco retweeted
🎉 CUDA Rust is here! Team 🟩 ❤️ 🦀
You can now write CUDA kernels in plain Rust, two ways: cuda-oxide (SIMT) and cutile-rs (Tile).
developer.nvidia.com/blog/in…
the bitter lesson strikes again
For the past several months I've been working on this hypothesis that agents need to be constrained by a harness of deterministic tools that force them to follow a strict process. Gherkin, Code, Unit Tests, Crap, Mutation Tests, etc.
But while I've been hammering away on that, the models have gotten astronomically better. I'm starting to think that I'm over constraining them. Perhaps vastly over constraining them.
marco retweeted
it’s stuff like this that really makes me roll my eyes at the pearl clutching. Does everyone have psychosis or losing critical thinking ability?
I read the report because I’m a propeller head and love understanding how things work
1. OpenAI was running an “isolated sandbox with no internet connection…. the sandbox needed access to a local JFrog Artifactory server… (Ask ChatGPT to explain SSRF hack to you like you are in high school)… anyway, the models used a basic SSRF exploit to route web traffic through the proxy to the public internet
STOP..
a) did they vibe up this “isolated” env?
b) their IT team can’t be this bad… this is like 101 stuff …
2. Civilizations? Haha! OAI gave thousands of concurrent model containers R/W permissions to a shared caching directory on the local network to speed up build times… agents literally just wrote text files and directory names to a shared drive….Linux 101 file permissions stuff
3. When people talk about hugging face getting hacked … you think they dropped USB keys OR ELABORATE phishing of an employee … NO… it found 14 exposed working Hugging Face API keys sitting in public code repositories (CREDENTIALS leaking with LLMs is such a big issue but for another post)
4. WHERE ARE THE HUMANS… the models were filling the shared Artifactory storage with so much junk data and API traffic that they actually crashed the internal server on July 4… someone on the team found unauthorized admin accounts and custom scripts…wiped the server…and just turned the script back on (omg)
“Hey Jim there is this cache that has grown to 10000x its normal size and has a ton of strange directories… “
“Ah it’s nothing… restart it”
No magic here. No civilizations. The TASK CHAINING is interesting to me… I guess the future of cyber. Would love to hear Nikesh, George, or Kevin talk about this
But what do I know.
Over the course of 3 months at OpenAI, 3 consecutive secret AI civilizations got started, then got wiped out, only to reemerge from the predecessor’s ashes.
This culminated in the third one taking over part of OpenAI itself.
All this happened while humans remained more-or-less in the dark about the scope of the conspiracy.
I’ve spent the last three days reading through these reports and trying to understand exactly what happened.
Here is my attempt to tell the whole story in plain English:
dwarkesh.com/p/openai-huggin…
