building software, rethinking systems, and enjoying life in chicago

Chicago, IL
Pinned Tweet
Writing bug reports to OSS projects shouldn't be quick. Maintainers are heroes for donating the amount of time they do to their craft, and you should allocate a modicum of yours to save as much of theirs as possible. Anything less comes off as obnoxious and entitled, IMO.
4
224
henry retweeted
Replying to @ThePrimeagen
AI allowing humans to do less robotic bullshit? Based. AI allowing humans to have fewer human interactions? Cringe.
1
3
219
Found a true monster playing #Witcher3 last night.
1
17
Much more value to be extracted from the things you're NOT doing than from the things you are. The latter is finite, but the former is infinite. So much is possible now. It's essential to re-evaluate your priors.
The thing people miss with local AI is that it doesn't need to replace your $200/mo frontier inference plans to be viable. There are tons of little automations that are super viable with small, open-weight models. The hard part is finding them.
10
God I am so hyped for overalls.dev More excited for this than GTAVI
😎 overalls.dev now supports pausing/resuming firecracker machines. here's how it works: if the machine is 'pinned' it shows up on the top of your dashboard and is prevented from being deleted for suspended. if the machine is created via /threads (ie. orb workflow) then the machine is automatically suspended after 4 hours of inacitivity from tool call. a new prompt/toolcall will unpause a machine. if a network request (ie. vscode web, ssh or shared webapp) arrives it and the machine is paused then it will automatically unpause/resume. if there is no capacity on the lilly node (the bare metal machine) then it will >automatically relocate the machine< to another lilly node with capacity machines can now be cloned (ie. forked) with their live memory state instead of being turned off (loss of memory state). of course you can pause/suspend/clone directly via ssh all of the above rules hold true even if the machine needs to be relocated to another lilly node with capacity. tldr; cpu/memory is expensive right now and now resource mgmt is built in as a platform primitive that you can self-host on your own infra 😎
1
82
henry retweeted
😎 overalls.dev now supports pausing/resuming firecracker machines. here's how it works: if the machine is 'pinned' it shows up on the top of your dashboard and is prevented from being deleted for suspended. if the machine is created via /threads (ie. orb workflow) then the machine is automatically suspended after 4 hours of inacitivity from tool call. a new prompt/toolcall will unpause a machine. if a network request (ie. vscode web, ssh or shared webapp) arrives it and the machine is paused then it will automatically unpause/resume. if there is no capacity on the lilly node (the bare metal machine) then it will >automatically relocate the machine< to another lilly node with capacity machines can now be cloned (ie. forked) with their live memory state instead of being turned off (loss of memory state). of course you can pause/suspend/clone directly via ssh all of the above rules hold true even if the machine needs to be relocated to another lilly node with capacity. tldr; cpu/memory is expensive right now and now resource mgmt is built in as a platform primitive that you can self-host on your own infra 😎
16
9
77
5,398
The thing people miss with local AI is that it doesn't need to replace your $200/mo frontier inference plans to be viable. There are tons of little automations that are super viable with small, open-weight models. The hard part is finding them.
21
I've built an incredible todo system with @liquidai D1-3B for action routing (e.g. create new or edit existing) LFM-2.5-24B-A2B for synthesizing transcription blobs into titles, descriptions, and deadlines LFM-2.5-Embedding-350M for semantic lookups Local AI at its best.
2
3
25
3,449
henry retweeted
Following Ukrainian strikes on Russian data centers: Tucker Carlson's views dropped 71% Douglas Macgregor's views dropped 62% TCN views dropped 34%
Following Ukrainian strikes on Russian data centers: Scott Ritter's views fell 43% Rick Sanchez's views fell 39% Max Blumenthal's views fell 37% Ben Swann's views fell a 18%
715
4,401
20,784
1,302,467
big if true
My current model rankings S - Opus 5.5, Muse Wonder A - GPT-6.1 Sol — Gap — Everything else
30
wut
Introducing Muse Wonder, our new family of frontier models with state of the art performance for personal and professional agentic work.
10
...or even otp-js 😱
OK being able to stick an entire runtime and agent development environment in a V8 isolate is dope. @elixirlang folks if TypeScript (or any language) is to "win" in the AI era, it'll be because of stuff like this. Legit think we might need an otp-rs... piped.video/ND_kRgU_WlY
14
OK being able to stick an entire runtime and agent development environment in a V8 isolate is dope. @elixirlang folks if TypeScript (or any language) is to "win" in the AI era, it'll be because of stuff like this. Legit think we might need an otp-rs... piped.video/ND_kRgU_WlY
Announcing tsc-rs (aka ts-rust), my complete rewrite of the Typescript compiler, type checker, and LSP, all in Rust. This is a project I've had agents working on for 5 months now. I burned ~$400k of Codex tokens and got nowhere. Burned ~$20k of Opus and got there in 2 weeks. I have not read a single line of the code. tsc-rs is an open source drop-in replacement for tsc, and it's available now.
57
"AI is going to revolutionize how we work and usher in a new age of Abundance for all." AI:
this is the most beautiful thing i've seen, EVER thank you so much Josie
15
My girl's yell hurts my ears. Shrill issue.
Pickles came out bad Dill issue
1
20
We’re releasing a broad range of new mathematical results produced by an internal frontier model. We’ve been consulting with the independent Advisory Group on Mathematics and Artificial Intelligence at the Institute for Advanced Study, and we have drawn on their advice and public recommendations to inform how we release these results. github.com/openai/math
26
🤪
Replying to @maria_rcks
Alpha increasingly comes from one's ability to break out of boring gradient descent and find novel local minima. Wacky ideas are the new moat.
9
Sheffield sold a QL1.2 today because of Instinct. Had a conversation a couple weeks ago about wanting a QL1 but they were sold out. Went from not knowing the QL1.2 existed to having one ordered for me in like four WhatsApp msgs. This is gonna be huge.
Instinct is the only one with a real shot IMO because they're leaning HARD into making it feel like texting a person, and also focusing on consumerism for their monetization strategy. Methinks the "ooh I want that" -> "here's a link" flow is gonna hit for a lot of people.
24
henry retweeted
Calvin was a true prophet of our time
We’re releasing a broad range of new mathematical results produced by an internal frontier model. We’ve been consulting with the independent Advisory Group on Mathematics and Artificial Intelligence at the Institute for Advanced Study, and we have drawn on their advice and public recommendations to inform how we release these results. github.com/openai/math
3
26
159
3,979
Friendly reminder that pushing the frontier of human knowledge for any reason is one of the safest bets you can make.
Replying to @elshayib_
It's one of those things where we don't know the utility of until we unlock the capabilities. NASA developed sophisticated computer algorithms to enhance pictures of the moon and then someone realized that they could be applied to the human body. And boom, MRI machines.
14