Live data from Hacker News

Viewing profile — tomdekan

tomdekan

HN member
Joined
Sat, Oct 16, 2021, 11:01 AM UTC
HN karma
155
Public activity
62 items

About tomdekan

I'm an AI and software creator.

https://tomdekan.com

Recent public activity

  1. comment
    Comment #45835462

    I've been enjoying using this over the past 3 weeks. This is nano-hydroxyapatite, meant to be more effective than hydroxyapatite alone. https://www.boka.com/ (P.S I'm not affiliate…

  2. comment
    Comment #45835450

    Xylitol reduces cavity‑causing bacteria and raises salivary flow, which can increase calcium and phosphate availability. That environment supports natural remineralisation and boos…

  3. comment
    Comment #44844545

    Post on Reddit showing many women as upset by OpenAI's new models. They viewed the previous models in ChatGPT as their boyfriends. With the new models, they now feel that their par…

  4. story
  5. comment
    Comment #44669472

    Thanks Daniel. Do you recommend any resources showing the differences between different quantizations?

  6. story
  7. story
  8. story
    Show HN: Simple AI-powered commit msgs script

    I wrote a short script that takes a Git diff and uses OpenAI’s API to output a concise commit summary plus a detailed description following best practices. It works with any repo -…

  9. comment
    Comment #43791158

    Also, I don’t think my previous comment does assume that there are compelling alternatives. A person always has a choice not to spend their money. Even if they need expensive healt…

  10. comment
    Comment #43791098

    I disagree with the general comment that “any established market” has “only a few entrenched players”. I’d say that most markets provide compelling alternatives. Where they don’t y…

  11. comment
    Comment #43780093

    That’s a very cynical view. Do the biggest companies not create the most value for the world? Consider this. If the most successful companies are simply cheating customers, then mo…

  12. comment
    Comment #43658529

    Do you have access to o4 models as your website suggests? I didn’t realise that they had finished training yet.

  13. comment
    Comment #43597024

    So, Quasar == Llama 4 Behemoth?

  14. story
  15. story
  16. story
  17. comment
    Comment #43464706

    Any more info on the new Google model?

  18. story
  19. story
  20. story
  21. comment
    Comment #41090171

    Completely agree. I understand that this is becoming more common in London. This is still a harsh practice.

  22. story
  23. story
  24. comment
    Comment #38775467

    Here’s a shameless plug for my collection of short, succinct Django articles at https://photondesigner.com/articles The latest article is: “The simplest way to build an instant mes…

  25. story