Live data from Hacker News

Viewing profile — sheepscreek

sheepscreek

HN member
Joined
Fri, Nov 04, 2022, 2:02 AM UTC
HN karma
2,425
Public activity
874 items

About sheepscreek

NOTE: I post a lot from my phone - so please excuse typos and other errors. Happy to discuss/clarify anything - just shoot me an email. ——-

Student for life. Live and breathe technology. Coding since '96. Almost everything interests me.

Special interests: AI, LLM, Econometrics, Sociology, Psychology, Programming languages.

Email: sheepscreek ÅT icloud D0T com

Recent public activity

  1. comment
    Comment #49191296

    It could be that they’re pitching Meta Muse 1.2 against Terra and Opus level models. They probably consider Sol to be a level above, along with Fable.

  2. comment
    Comment #49181087

    I couldn’t help but feel a bit emotional reading this. Elise was clearly an incredible human. Thank you Stephen for sharing her story and your love for her. I cannot imagine a bett…

  3. comment
    Comment #49154112

    I was super excited to read this but lost the plot when I go to > Qwen3.8-Max was asked to create the oh-my-cli project from scratch and, over a 10+ day long-horizon autonomous cod…

  4. comment
    Comment #49135957

    I didn’t imply it was novel.

  5. comment
    Comment #49135225

    High quality research - my mind is blown by the concept of replacing “reasoning” with “…” (literally) with no effect on actual output.

  6. comment
    Comment #49135126

    It gives every employee their own personal assistant (AI agent). It learns their preferences, and more importantly, can access whatever they can.

  7. comment
    Comment #49046614

    I don’t think this is true if a disclaimer is provided. Most investment advice videos on YouTube make this very clear, but it doesn’t stop them from sharing advice. I suppose there…

  8. comment
    Comment #49031732

    You're good. Dude that topped Meta's tokenmaxxxing board before it was shut down used 265 billion tokens in a month. I kid you not.

  9. comment
    Comment #49027035

    I really hope the lawsuit does not end in OpenAI tightening the “safety” and “security” of their models into oblivion. I hope they understand that a lobotomized model is as good as…

  10. comment
    Comment #48951798

    Hey Thariq. Don't dwell too much on the negativity here. It's a hard act to balance and I'm sure the frightening pace has only made your job harder. I think there might be a case f…

  11. comment
    Comment #48940395

    AFAIK there’s no ZDR with Claude models accessed directly via Anthropic. You’d have to go through either Google Vertex, Azure or AWS for true ZDR (at least legally/on paper).

  12. comment
    Comment #48921074

    I’ve had great luck building SwiftUI apps with GPT-5.5 (now GPT-5.6 Sol), Opus 4.8 and Fable 5. I’m just offering another data point, not suggesting on the effectiveness of DSLs fo…

  13. comment
    Comment #48905610

    Agreed. I think we’re entering an era where some level of specialization for general LLMs is a good thing. Particularly between tuning for agentic use cases (where you want agency …

  14. comment
    Comment #48848963

    I wonder if they are double-counting Anthropic's leased capacity from SpaceX under SpaceX again.

  15. comment
    Comment #48848854

    I don't think it even matters. Because noone will continue to use an LLM that doesn't work well for them, whether or not it has a good bench result. So for their own sake, the corr…

  16. comment
    Comment #48747825

    Moneyshot: > The three bacterial strains that successfully induced tumor regression (E. americana, C. portucalensis, and E. ludwigii) were all identified as facultative anaerobic b…

  17. comment
    Comment #48740733

    Yes - they create new/cheaper products for a different consumer but not at the cost of their margins. Vision Pro may have been the only device in recent history that likely had sli…

  18. comment
    Comment #48736517

    > It's easy to blame datacenters, but there are many factors at play here. Henrico County currently has 37 data-centres with ~2 gigawatts capacity (expected to reach 3 gigawatts). …

  19. comment
    Comment #48687487

    I’m a complete layman to this field, but what the article did say was they’re hopeful that AI/ML can help develop a model that can pull out information such as the scattering cause…

  20. comment
    Comment #48686027

    The message I’m getting is that Apple will never compromise on its healthy margins. If something becomes basically unaffordable for their target market, they’d cut the production a…

  21. comment
    Comment #48667694

    Speculating here - “effectively” cooling the CPU and GPU materially using this technique at datacenter scale may have never been done. Those things than run hot, easily crossing 10…

  22. comment
    Comment #48667576

    I think it’s more than that. Piecing together the perspective of a few commentators in this post - it’s plausible Anthropic is trying to shift the narrative from US vs. Rest of the…

  23. comment
    Comment #48646484

    So I think the takeaway here is, this is a super fast companion model to larger models, that reasons quickly. Perhaps this technique can be used to train a highly optimized reasoni…

  24. comment
    Comment #48639233

    The initial motivation for this was likely to thwart any competition. Already Anthropic has accused some companies of organized distillation efforts at a massive scale. Back when I…

  25. comment