Next time you're vibe coding something, have the system generate a mermaid diagram to show its understanding. Though visual generation can be hard for models, structure/topology in formats like mermaid is pretty gettable.
I've even found sonnet and opus to be quite capable of generating json describing nodes and edges. I had them generate directed acyclic processing graphs for a GUI LLM data flow editor that I built (also with Claude - https://nodecul.es/ if curious)
BMW and Toyota have famously used bio-derived insulation reported to be like catnip for rodents.
The bio-oil plasticizers also migrate out more quickly in thermal cycling than the old dead dinosaurs approach. Hilariously, when I asked my mechanic about getting an M5, he laughed and explained that the radiator components are known to turn brittle and crack after 5-6 years because of this.
(I don't envy automotive folks. The stuff they have to deal with is next level.)
Last time I had to call AAA to jump my car, the guy opened the hood very carefully and told me he’d had three rats jump out of engines at him that day, presumably because of the “soy wires.”
Having come from graphics in the 90's, practical high-performance answers typically involve fakery on both primary surface shading and shadow calculation.
I've pulled some tricks like "object-pre-pufficiation" (low-frequency model manifold encapsulation, then following the same bones for deformation) mixed with normal recording in shadow layers (for realtime work on old mobile hardware), but, these days, so much can be done with sampling and proper ray-tracing, the old tricks are more novelty than necessary.
If you have reliability and accuracy (big if) then the practical usability and cost become performance problems.
And this is a bit of a sliding scale. Of course users want the best possible answer. However, if they can get 80% (magic hand-wavey fakie number) of the best answer on one second instead of 20, that may be a worthwhile tradeoff.
I don't know if there's a named law, but the word for not knowing and believing that something remembered is a novel idea is "cryptomnesia".
Knowing that you know something by teaching is Feynman's method of understanding. Basically, on scanning, I don't particularly disagree with the content of the post. However, treating these things (many of which regularly show up here on HN) as being due to "14 years at Google" is a little misplaced.
But, hey, it's 2026, CES is starting, and the hyperbole will just keep rocketing up and out.
Huh... they better build some readily-available hyper-powerful infrastructure, pronto, or that next election could hand power to folks that don't have the best interest of the country in mind:
Aggregability is NP-Hard... Useful the next time someone insists that it's possible to find a "perfect" model for a non-trivial ML problem. (I get this ask 1-2 times per month.)
I had the opportunity to meet with folks from Histosonics at a Canopy Cancer Collective (pancreas cancer focused group - https://canopycancer.org/) annual meeting a couple of years ago. They had shown very promising results (and approval) with liver cancer, and the applicability to any soft-tissue openly-addressable masses (e.g., not brains in skulls, not lungs full of air) seemed very likely, based on the physics. (Note: I'm a consumer electronics and ML engineer, not a medical devices engineer).
I'm excited to see this option become more broadly available. The ability to precisely target and illicit an inflammatory response is impressive, and Whipples are no joke.
I've even found sonnet and opus to be quite capable of generating json describing nodes and edges. I had them generate directed acyclic processing graphs for a GUI LLM data flow editor that I built (also with Claude - https://nodecul.es/ if curious)
reply