Live data from Hacker News

Viewing profile — CuriouslyC

CuriouslyC

HN member
Joined
Wed, Jul 30, 2014, 2:03 PM UTC
HN karma
8,051
Public activity
4,582 items

About CuriouslyC

Chief scientist/ceo, Sibylline Software. nathan @ sibylline.dev

Recent public activity

  1. comment
    Comment #49246621

    Unlike 100 years ago, the bald thing is only a negative if you're fat/weak looking. The bald tough guy has been so established in modern lore that even older men can ride the wave …

  2. comment
    Comment #49246526

    My experience is for men that it's not just white collar/blue collar, more it's appearance gives a halo that helps people progress in domains where there's a human, interpretive as…

  3. comment
    Comment #49234577

    I've pivoted to game dev because I'm burned out on AI dev tools. I started working on a hexpunk skirmish roguelite deckbuilder (working name is Hexborne) where you place units and …

  4. comment
    Comment #49222333

    Blue Prince immediately came to mind for me as relevant modern art. Honestly Myst style puzzle games seem like the easiest type of game for a generative world model to simulate, so…

  5. comment
    Comment #49218347

    Agents are getting pretty good at Nix. I think it has sufficient critical mass that it'll be fine.

  6. comment
    Comment #49214213

    Kotlin my dude, unless you hate yourself.

  7. comment
    Comment #49209525

    Trump is either going to break the US, or the Republican party. Either way, we're in for a long period of one party rule and chaos.

  8. comment
    Comment #49209506

    If China is declining more slowly than the rest of the world, that's still relative progress.

  9. comment
    Comment #49209482

    They even look like Captain Planet baddies. Trump with his orange skin, purple sores and ridiculous hair, RFK jr. walking around barefoot looking like a scrotum on steroids, Pete K…

  10. comment
    Comment #49202608

    That's not the only reason to go to expert models. The more different domains you try to stuff in there, the more parameters the model needs to keep things coherent and not overloa…

  11. comment
    Comment #49202529

    Ironically, Opus 5 is the most benchmaxxed model I've seen from Anthropic. It is legitimately smart in a lot of ways but it has communication issues, both in terms of how it commun…

  12. comment
    Comment #49168600

    [flagged]

  13. comment
    Comment #49168512

    My bias is that I'm tired of people being mad at Musk/Zuckerberg/Altman/Amodei for being jerks (undisputed), and corrupt politicians allowing data centers to be put where they real…

  14. comment
    Comment #49168426

    It does more than that, it shames him for being lazier than the people he's attacking.

  15. comment
    Comment #49168400

    [flagged]

  16. comment
    Comment #49168368

    [flagged]

  17. comment
    Comment #49168312

    It'd be nice if blog authors categorized their posts into "old man shaking fist at clouds" vs "something I searched for on the internet, legitimately never found it, and decided to…

  18. comment
    Comment #49168270

    Hence the advice to read the tests? Interesting how people make the argument that AI code is garbage and can't be trusted (it's inelegant and poorly structured, sure, but that's a …

  19. comment
    Comment #49168191

    [flagged]

  20. comment
    Comment #49168147

    [flagged]

  21. comment
    Comment #49168134

    Plot twist, 100% likely you've read and shared humanized AI content. If you read the summary/abstract and nope out because there's nothing interesting or insightful, whatever, but …

  22. comment
    Comment #49168051

    [flagged]

  23. comment
    Comment #49168022

    Yet the kids can't stop doing either... Which was the OP's point.

  24. comment
    Comment #49167903

    [flagged]

  25. comment
    Comment #49167876

    Human slop is slop too.