Hacker News: throwaway63467

New comment by throwaway63467 in "Why are coding agents so dumb?"

throwaway63467 — Fri, 09 Oct 2026 22:06:04 +0000

Same, it’s so smart but so dumb. I explain an architecture change to it and write examples of strongly typed Go code and how to store structure, it agrees and then proceeds to write some untyped string map ball of mud that has 900 specializations.

But maybe that’s on us, AI doesn’t care about all these special cases, it’s not debt to it as it will simply read them all when making changes. We’re obsessed with quality and what code is supposed to look like but those are human standards, AIs evolve to look at this complexity as a single picture, they can simply see through it so what is spaghetti code to us is merely some code to them that works as it should and is efficient. It’s interesting we can see how the two things drift apart, you would think at some point AI generated code should explode but it hold together unreasonably well in most cases…

New comment by throwaway63467 in "Nicholas Polson has authored 258 academic papers in 2026 so far"

throwaway63467 — Tue, 29 Sep 2026 20:00:38 +0000

I tried writing a simple 20 page paper with the latest Claude model over several weeks and that thing got ripped apart by legal and data science (for good reason), not going to try that again. And there I had all the numbers and analyses down and verified, just the writing itself by Claude was subpar.

Can’t imagine the utter garbage that Claude will produce when you give it a simple prompt and let it write thousand of pages.

Kind of makes me wonder about all that AGI talk if these models can’t even the work of a good PhD or postdoctoral student.

New comment by throwaway63467 in "CAPTCHAs don't prove you're human – they prove you're American"

throwaway63467 — Sat, 26 Sep 2026 09:20:08 +0000

Doesn’t recaptcha have localization? Not sure now that I think about it maybe not

New comment by throwaway63467 in "The Mafia may be keeping fentanyl out of Italy"

throwaway63467 — Fri, 25 Sep 2026 12:46:02 +0000

I guess if you’re the mafia you just murder the competitors. Can’t have stronger regulation than that!

New comment by throwaway63467 in "Radicle: Disclosure of Vulnerability in the Network Protocol"

throwaway63467 — Wed, 23 Sep 2026 22:29:53 +0000

Just use mTLS via QUIC, it’s standardized, secure, future proof and has implementations in tons of languages and supports proper certificate checks and that whole ecosystem around it which they apparently tried to reinvent. It’s such a great protocol for these use cases, I don’t get why it’s not used more.

New comment by throwaway63467 in "I built non-autoregressive decision models with RL a year ago"

throwaway63467 — Sat, 19 Sep 2026 14:23:51 +0000

Landing page full of AI fluff, discussion feels very fake here, I would assume this is some upvote bot, nothing makes sense.

New comment by throwaway63467 in "CCC invites all model citizens to 40C3"

throwaway63467 — Thu, 17 Sep 2026 17:16:35 +0000

Lots of families there every year and older folks too, it’s actually a great date for a lot of IT folks as there aren’t many weeks when companies are effectively shut down, maybe even only this one week.

Where's My Alien Mind?

throwaway63467 — Wed, 16 Sep 2026 17:37:21 +0000

If we are about to achieve the AI rapture and models are solving Millennium Prize Problems, where is my AGI??? I just tried for weeks to write a simple paper with Claude and even though I put a ton of effort in context engineering, material collection and prompting it degraded quickly into an unintelligible mess of a word salad. Mind you that was with Fable 5.1, a model that supposedly is so sharp that it’s under export ban in the US for certain domains. Kind of hard to fathom seeing it fail so hard on a simple task that I was then able to do manually in the end.

So let’s stop pretending these things are going to replace any serious knowledge work in the coming years, I don’t think they can make them that much smarter for such tasks as I can’t see any difference between the latest model updates, make them write anything more than a few pages of text in a novel domain and it sounds like a stochastic parrot. And I have tried so many strategies to make it reflect on what it writes, it just doesn’t have the ability to think like a human can, it just simulates thinking to a degree that is convincing while you only look at a small sample of its output. I found it’s the same with code, small changes work reasonably well and you can’t see the defects but keep doing that for a few months and you’ll have stochastic code soup. Not sure if these things make me even 2x more efficient in the long run for producing new stuff. They make me faster for search and lookup, but not anything that requires thinking beyond the simplest mechanical form. And it doesn’t help me that it can solve a Millennium problem, which btw it can only do while being guided by a team of world class mathematicians. So it’s at best an ability multiplier that’s tethered to the human intelligence steering it.

Comments URL: https://news.ycombinator.com/item?id=49730388

Points: 3

# Comments: 1

New comment by throwaway63467 in "There Is No AI (It's Just People) with Jaron Lanier"

throwaway63467 — Mon, 14 Sep 2026 12:50:25 +0000

They will settle for what’s available at the time, I dont think it makes sense either but they will come up with something that is good enough for their purposes and as long as the copy parrots the original well enough it will pass.

New comment by throwaway63467 in "There Is No AI (It's Just People) with Jaron Lanier"

throwaway63467 — Sun, 13 Sep 2026 21:01:16 +0000

The more interesting question is what happens once we replace manual labor with artificially intelligent robots and we arrive at a place where 99 % of human labor got eliminated. How do you maintain civil order? Every country is built on the premise that you have to “earn your keep” but what when that’s no longer true? Think about it, no labor, only capital. What will the people owning the capital demand of people that don’t own any? They won’t be useful to them in any way. I expect something like forced reduction of fertility for regulars to depopulate countries, while hyper rich individuals will extend their life spans and become immortal. And I don’t mean they will really survive but someone like Elon Musk will have their consciousness mapped to an AI and will instate that AI as their official heir. That sounds crazy but they will totally go for it just listen at what Peter Thiel is spouting in terms of immortality and becoming godlike. Hope people will have enough left to say to prevent that but right now it feels we’re going hardcore towards that scenario.

New comment by throwaway63467 in "David Sacks: OpenAI and Anthropic Don't Need Regulations to Pace Frontier Models"

throwaway63467 — Sun, 13 Sep 2026 17:44:06 +0000

I guess what they really want is to limit sale of AI models to compliant vendors and then raise the bar to compliance just high enough so they can pass it but smaller labs can’t. There’s no moat, it’s an efficient market that drives margins to zero right now, of course they don’t want that, collusion of the big vendors is the next logical step.

New comment by throwaway63467 in "Libraries Run Rust Inside Python (With PyO3)"

throwaway63467 — Sun, 13 Sep 2026 16:52:37 +0000

There’s a whole legacy of wrapper interface generators like SWIG that are way more powerful than PyO3, QT is e.g. fully mapped to Python using a very powerful wrapper generator (SIP). PyO3 is just a really simple interface generator in comparison, look at what SWIG can do, or SIP, what PyO3 does is trivial in comparison.

New comment by throwaway63467 in "Libraries Run Rust Inside Python (With PyO3)"

throwaway63467 — Sun, 13 Sep 2026 16:44:32 +0000

They’re not running Rust they’re running machine code compiled with the Rust compiler. Like many other Python libraries such as numpy are running machine code compiled with C++, C or Fortran. That was always the case for Python and is its main selling point, it’s a slow but easy to write glue language that can make faster native code scriptable.

Also the article reeks of AI slop, it’s just trying to sell you a Rust course.

New comment by throwaway63467 in "Don't be the out of touch Kung Fu master"

throwaway63467 — Sun, 13 Sep 2026 15:46:58 +0000

But he didn’t build the Oculus did he? Not sure what he worked on, software?

New comment by throwaway63467 in "Astra for Coding: Why Are We Doing This Again?"

throwaway63467 — Fri, 11 Sep 2026 09:29:30 +0000

Also my experience, it works somewhat ok on large code bases that I designed and built myself before but after months of agentic development they sure start to degrade. I think if you start from scratch with agentic development there is no foundation for the models to anchor to.

New comment by throwaway63467 in "More questions about whether researchers can trust OpenAI with unpublished math"

throwaway63467 — Thu, 10 Sep 2026 19:34:08 +0000

Isn’t that the whole spiel of these things, you run all kind of text and other data through it and it kind of remembers it and learns from it then it spouts it back out like a human would. Makes sense to me that a training run based on conversations that were fed into the system by users is results in the model learning from these so the model will spit the knowledge back out again, just in a way that’s not directly attributable to the original content (which is the most important step as otherwise it would just be plagiarism). I guess that’s why OpenAI can get better and better as well so fast, people work with it and teach it how to do things by giving it feedback and iterating with it, and all that goes back into the training loop. And training data about millennium prize problems is probably quite spars. Wonder if anyone has tried injecting nonsense science into the training data (e.g. work out a fantasy science theory with names and all kinds of stuff) to see if the model will regurgitate it in a couple of months for other users.

New comment by throwaway63467 in "How GPT‑5.6 Sol helps run quantum computing experiments"

throwaway63467 — Wed, 09 Sep 2026 08:12:47 +0000

I had qubit bring up and calibration fully automated with Python in 2011 including full spectrum measurements, lifetime characterization, Rabi/Ramsey measurements, calibration of single qubit gates and two qubit swap gates and full quantum process tomography, so not sure if AI is really needed there, curve fitting and some data logging is enough for this. Even had a nice LabView like GUI but with PyQT, it was quite nice. Still of course cool, I guess today I would just let Codex loose on some experiment goals but in the end my ability to produce results was mostly limited by the chip itself and the qubit lifetimes and theres no magic trick AI can apply to make these go up by a factor of 10. Still would’ve saved me a lot of time for routine programming tasks I imagine and that seems to be the main takeaway of the article. I guess name dropping quantum computing makes this sound cooler but in principle it’s just automation that you can apply anywhere, nothing quantum computing specific here.

New comment by throwaway63467 in "The Navier–Stokes Millennium Prize Problem"

throwaway63467 — Wed, 09 Sep 2026 07:03:35 +0000

Well OpenAI wants to maintain its edge and solving a Millennium problem is of course fantastic PR, and given that Anthropic seems to be working on that it’s not hard to imagine they wanted to be first. I don’t get how people buy into that whole “oh we heard models can solve Millennium prize problems now so we thought why not give it a go…” story - it’s a bit funny. This whole AI bubble is about hype and solving such problems is probably one of the best ways to keep this hype up so you can safely assume both OpenAI and Anthropic are using significant resources on these areas.

New comment by throwaway63467 in "The Raspberry Pi Interactive Timeline · 2006–2026"

throwaway63467 — Sat, 05 Sep 2026 09:52:06 +0000

The USB C connector has like 24 tiny SMA pins how is that trivial? Under the microscope with soldering equipment from a lab maybe, at home this seems impossible and it will likely not work due to the high frequencies involved, I don’t think you can get the impedance right when hand soldering this back on.

New comment by throwaway63467 in "The Raspberry Pi Interactive Timeline · 2006–2026"

throwaway63467 — Fri, 04 Sep 2026 09:32:40 +0000

Broke off my Pi 4Bs USB C port when using it to test link speed for new LAN outlets, when I wanted to reorder I realized the 4 GB version is now 120 EUR… Damn.