Jake retweeted
masked decoding on gemma 3 12b to control text rivers
if a token isn't the length i want it to be, i mask it out
added a lora for width tags so the model is aware of text length + coherence RL afterwards
Jake retweeted
got to help Cooper Union's School of Architecture make a new website to highlight student work, rotating links of community events, and a collaborative playground to build with common materials (using playhtml ofc!)
check it out on cmu.place :)
Jake retweeted
Drag-and-drop makes the digital graspable, but building it well is fiddly enough that many don’t bother. Declarative Dragging (with Josh Horowitz & Jeff Heer, #uist2026 best paper award), lets you make an element draggable just by listing the states dragging it can reach.
Introducing Kino: make videos end-to-end with Claude Code and Codex.
Drop your footage, get highlights, edit together, send links for feedback, and let your agents cook.
AI is becoming a great creative partner. Kino is a new home for your team and agents to collaborate.
Jake retweeted
New Anthropic research: How AI agents shape experts' judgment calls
When experts delegate work to an agent, do they still make the key decisions? We tested this in a randomized trial with 50 professional data analysts.
Jake retweeted
i got spooked about Instinct's privacy so i rolled my own - biscotti is a lightweight (~850 LOC) iMessage agent built on off-the-shelf tools (Pi, Firecracker VMs, Nix, Tailscale) to support computer use.
i'm daily driving it on my Raspberry Pi and don't miss anything!
Jake retweeted
interpretability might be the most important scientific problem in the world right now, as I don't think we can solve alignment if we don't know what we're building. I think I have good news about it. I co-authored early work in mechanistic interpretability at OpenAI and came away pessimistic, because neural networks appeared to fold into arbitrary tangled knots. years later, I started openai's first large-scale automated interpretability project, finding about 2% of neurons legible with the primitive tools we had
with modern agentic tools and a much closer look, the picture looks totally different. the vast majority in models up to trillions of parameters are legible, and their circuits beautiful. many famous examples of tangled neurons turned out to be misunderstood. The roadblocks I expected mostly aren't there. I now think ambitious interpretability will be solved by the time we need it most, and possibly quite soon, as automated scientists keep improving. it may be a bitter lesson for interpretability: the simple thing, at absurd scale, works
already-clean.goodfire.com/
Jake retweeted
My dear friends, I’d like to show you my latest toy project
Introducing Handwired: turn your regular JavaScript code into a neural net!
- no training
- exact outputs vs the JS program
- fully differentiable
All runnable in a browser, without GPU
Jake retweeted
open-sourced 160+ sound visualization experiments across 25+ categories. waveforms, spectrums, matrix displays, particles, and rhythm. all made while building iwrzwr, a sound app coming soon.
play with them live, steal the code, make them yours 💗
kagan.in/iwrzwr/visual-archi…













