Hacker News: aixpert

New comment by aixpert in "Not all elementary functions can be expressed with exp-minus-log"

aixpert — Wed, 15 Apr 2026 06:52:18 +0000

tangenially it would be interesting to analyze what infinite structures can be represented with infinite EML trees

New comment by aixpert in "All elementary functions from a single binary operator"

aixpert — Tue, 14 Apr 2026 14:47:27 +0000

this should be a tangential proof for the dying bunch of people who still believe that LLMs are just parrots. EML are literally a new invention

New comment by aixpert in "I’m joining OpenAI"

aixpert — Mon, 16 Feb 2026 06:37:49 +0000

he likely poured oil into the flame investing a few hundred bucks to double the virality

New comment by aixpert in "AI fatigue is real and nobody talks about it"

aixpert — Mon, 09 Feb 2026 07:48:19 +0000

as long as it's new I tremendously enjoy binge watching Claude: I have three tabs open and if one of them is not doing something interesting I just switch to a different channel, and occasionally influenced the narrative

New comment by aixpert in "The first sodium-ion battery EV is a winter range monster"

aixpert — Mon, 09 Feb 2026 07:42:33 +0000

you deserve a high value metal medal

New comment by aixpert in "Experts Have World Models. LLMs Have Word Models"

aixpert — Mon, 09 Feb 2026 07:36:07 +0000

The article basically claims that LLMs are bad at politics and poker which is both not true (at least if they receive some level of reinforcement learning after sweep training)

New comment by aixpert in "Experts Have World Models. LLMs Have Word Models"

aixpert — Mon, 09 Feb 2026 07:27:57 +0000

you are living in the past these models have been trained on image data for ages, and one interesting find was that even before that they could model aspects of the visual world astonishingly well even though not perfect just through language.

New comment by aixpert in "Experts Have World Models. LLMs Have Word Models"

aixpert — Mon, 09 Feb 2026 07:25:55 +0000

they do model the world. Watch Noble price winner Hinton or let's admit that this is more of a religious question then the technical.

New comment by aixpert in "Experts Have World Models. LLMs Have Word Models"

aixpert — Mon, 09 Feb 2026 07:23:01 +0000

you are correct the token representation gets abstracted away very quickly and is then identical for textual or image models. It's the so-called latent space and people who focus on next token prediction completely missed the point that all the interesting thinking takes place in abstract world model space.

New comment by aixpert in "Ask HN: Why all the sudden people are writing browsers with AI?"

aixpert — Wed, 28 Jan 2026 07:36:50 +0000

someone might build enough hype around it to challenge the oligopoly.

As for training on existing browsers: they are trained on the whole corpus of the human thought process and can take many insights from other fields into the browser, they can write a browser in a completely new language without just transpiring but building it from first principles or as karparthy calls it spec based programming.

New comment by aixpert in "A few random notes from Claude coding quite a bit last few weeks"

aixpert — Wed, 28 Jan 2026 00:06:44 +0000

rust compiler and redox operating system with modified Qemu for Mac Vulcan metal pipeline ... probably not junior stuff

you might think I'm kidding but Search redox on github, you will find that project and the anonymous contributions

New comment by aixpert in "Clawdbot - open source personal AI assistant"

aixpert — Mon, 26 Jan 2026 07:20:22 +0000

this is the whole message of this hype that you can churn out 500 commits a day relatively confidently the way you have clang churn out 500 assemblies without reading them. We might not be 100% there but the hype is looking slightly into the future and even though I don't see the difference to Claude code, I tend to agree that this is the new way to do things even if something breaks on average it's safe enough

New comment by aixpert in "Clawdbot - open source personal AI assistant"

aixpert — Mon, 26 Jan 2026 07:16:16 +0000

what still doesn't click for me is what can it do that Claude can't do

New comment by aixpert in "Clawdbot - open source personal AI assistant"

aixpert — Mon, 26 Jan 2026 07:14:10 +0000

these verbs seem appropriate when you accept neutral (MLP) activation as excitement and DL/RL as decision processes (MDPs...)

New comment by aixpert in "Claude Chill: Fix Claude Code's flickering in terminal"

aixpert — Wed, 21 Jan 2026 03:54:31 +0000

except when it needs to do tasks then it's the slowest of them all

New comment by aixpert in "Nanolang: A tiny experimental language designed to be targeted by coding LLMs"

aixpert — Tue, 20 Jan 2026 07:07:02 +0000

there was nothing nano about that language at all if you want true nano drop return keywoards or just go APL

New comment by aixpert in "Apple picks Gemini to power Siri"

aixpert — Tue, 13 Jan 2026 07:16:22 +0000

it's not going to help them. For Siri to be really useful it wouldn't need deep system integration and an external model is not going to provide that. People don't believe me when I said it about Apple Intelligence with open AI

New comment by aixpert in "Cowork: Claude Code for the rest of your work"

aixpert — Tue, 13 Jan 2026 06:39:10 +0000

claude

New comment by aixpert in "Cowork: Claude Code for the rest of your work"

aixpert — Tue, 13 Jan 2026 06:25:58 +0000

snapshots are local Time Machine backups for a few hours which don't need external hard drives and are configured by default I think

New comment by aixpert in "Google CEO Pushes 'Vibe Coding' – But Real Developers Know It's Not Magic"

aixpert — Sat, 29 Nov 2025 07:17:39 +0000

i'm a real developer and it is absolutely magic! if someone showed me a demo of talking to a computer which works for one hour to find and fix my bugs all on its own five years ago I'd definitely call it magic and I still call it magic.

does it make a mistakes? yes sometimes but you can verify with tests or with lean.