David Z retweeted
there's something really cool about asking your agent to draw a statechart (hypothetical example in this image), then using the /create-formal-verification-skill you made with pstack to find bugs through something like TLA+. you can both understand your code better and how the user flow should work, and fix issues users would otherwise see
pstack v0.15.27 ships with a new skill /create-formal-verification-skill!
it's still in beta, and similar to /create-verification-skill, this is a meta-skill that creates a new skill and formal verification suite based on the characteristics you want to verify. for example, you may want to model state machines and systems with concurrency using TLA+, or use kani in rust to check for undefined behavior.
if you're new to formal verification, think of it as mathematically proving properties and invariants (something that must always be true) or "laws" of your system. usually, this means modeling your programs in a separate language like with TLA+ and then running the solver on that spec. newer approaches like with kani and @bendlang take a different approach where there is no separate language: you write kani in rust, and bend in bend (it's also a programming language, and the proofs are written as laws and expressed in its type system).
historically formal verification never reached mainstream status because of the difficulty in modeling and writing proofs. it's quite difficult to model your business requirements and keep them in sync! but i expect that we will see more innovation in formal verification in the future as being able to formally verifiy that the code does what you intended gives your agents superpowers in writing correct software.
in the meantime though, i've been experimenting with modeling our code with TLA+ and have already found several bugs with the skill created by /create-formal-verification-skill.
try it out and let me know how it goes!
David Z retweeted
🚀 Introducing DeepInfer, our in-house inference framework, now fully open source!
Powering DeepSeek-V4.1-Flash at scale with high throughput and low latency. We’re sharing our production stack to help the community lower compute costs, reduce energy use, and make AI more affordable.
🔗 github.com/deepseek-ai/DeepI…
1/4
David Z retweeted
Le physicien Paul Dirac (ici à côté de Robert Oppenheimer) est mort il y a 42 ans.
Indifférent au froid, à la pluie, à l'inconfort, à la mauvaise qualité de la nourriture, il aurait pu travailler sur une île déserte.
Il vivait dans une sorte d'ascétisme, comme s'il avait voulu que son âme fût délestée des besoins du corps, qu'elle préservât sa nature afin de parvenir dans « l'autre monde » prête à s'unir à ce qu'elle y contemplerait.
Aussi long que mince, il éprouvait une sorte de dédain diffus pour les choses matérielles, les plaisirs ordinaires, les modes, les balancements de jupe, les bruissements de buvette. Jamais il ne fuma ni ne toucha une goutte d'alcool. Un Gandhi en costume trois pièces, le militantisme en moins.
Mais cet homme flegmatique fut un chercheur fulgurant et prolifique. Sa vie fut certes rectiligne et banale, mais ses pensées, elles, se développèrent avec une intensité très particulière. En dépit de sa réserve, de sa crispation puritaine, il allait, à sa façon, découvrir et explorer un tout nouveau continent : l'antimatière.
Il passa aussi beaucoup de temps à s'occuper de son jardin. Il tenta de traiter les problèmes d'horticulture en partant des premiers principes de la physique (conservation de l'énergie, croissance de l'entropie...). Les résultats ne furent pas à la hauteur de ses espérances.
David Z retweeted
„Koń będzie zawsze istniał, ale samochód to tylko chwilowa moda – zabawka dla bogatych.”
Michigan Savings Bank, 1903
„Myślę, że na całym świecie jest rynek dla może pięciu komputerów.”
Thomas Watson, 1943
„Żadnej szansy, że iPhone zdobędzie znaczący udział w rynku.”
Steve Ballmer, 2007
„Internet ma dla gospodarki ledwie ważniejsze znaczenie niż faks.”
Paul Krugman, 1998
„Maszyny latające cięższe od powietrza są fizycznie niemożliwe.”
Lord Kelvin, 1895
„Problem spamu w e-mailach zostanie rozwiązany w ciągu maksymalnie dwóch lat.”
Bill Gates, 2004
„Telewizor nie utrzyma się na rynku. Ludzie szybko znudzą się wpatrywaniem się co wieczór w skrzynkę ze sklejki.”
Darryl F. Zanuck, 1946
„Nie ma powodu, dla którego ktoś chciałby mieć komputer w swoim domu.”
Ken Olson, 1977
„Kiedy paryska wystawa światowa się zamknie, światło elektryczne zgaśnie i więcej o nim nie usłyszymy.”
Erasmus Wilson, 1878
„Amerykanie może potrzebują telefonu, ale my nie. Mamy wystarczająco dużo posłańców.”
William Preece, 1876
„Bez względu na to, jak nauka będzie postępować w przyszłości, człowiek nigdy nie dotrze na Księżyc.”
Lee De Forest, 1957
„Podróżowanie z dużą prędkością jest niemożliwe, ponieważ pasażerowie przestaną dostawać powietrza i umrą z powodu uduszenia.”
Dionysius Lardner, 1830
„Zabawa z prądem zmiennym to czysta strata czasu. Nikt nigdy go nie wykorzysta.”
Thomas Edison, 1889
„Telefony komórkowe absolutnie nie zastąpią telefonów stacjonarnych.”
Marty Cooper, 1981
David Z retweeted
As much as I respect Chinese-made AI models and hardware, I'm equally puzzled by how rough most of the software still is. Whether it's Wechat or Alipay or HarmonyOS. You'd think getting unique aesthetics, perfect fidelity, and great flows would be "easy", but clearly it is not.
David Z 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/
