Live data from Hacker News

Viewing profile — clauderoux

clauderoux

HN member
Joined
Sat, Jul 20, 2019, 11:47 AM UTC
HN karma
208
Public activity
125 items

About clauderoux

No profile information was provided.

Recent public activity

  1. comment
    Comment #48457697

    I have been a pretty consistent user of AI since 2022 (Instruct-GPT), so I don't have a bad opinion about the topic. However, I think the real problem now has become pretty obvious…

  2. comment
    Comment #48119341

    LispE has been fully ported to WebAssembly, you can see it in action with some examples.

  3. story
  4. comment
    Comment #48035521

    LispE is a programming language that has been ported to WASM. It's lightweight (about 3.3MB) and provides many features natively, with no external dependencies. You can now try it …

  5. story
  6. story
  7. comment
    Comment #44808682

    This article nails it. The claim 10x is in my opinion one of these tactics used by large corporations to force engineers into submission. The idea that you could be replaced with a…

  8. comment
    Comment #44795341

    Formulas can be written in English and translated by an AI in JS formulas.

  9. story
  10. comment
    Comment #44743685

    I love how so many people are eager to criticize LLM code, when in fact, according to my experience it is pretty superior to anything I have seen produced by human programmers, mos…

  11. comment
    Comment #44720301

    The answer is very easy. Latinate words (mostly of French origin for the matter), first because of the number of people speaking romance languages (French, Italian, Spanish, Romani…

  12. comment
    Comment #44692371

    Funny as they always seem to forget the hardware side of search engines. Google was incredibly fast compared to its competitors because they were among the first to store their who…

  13. comment
    Comment #44569626

    Well. There is measles epidemic right now...

  14. comment
    Comment #44540758

    Some people still think that LLM are just word predictors. Technically, it is not. First, transformer architectures don't process words, they process semantic representations store…

  15. comment
    Comment #44478743

    For those who don't know, "frelon" is French for hornet.

  16. comment
    Comment #44411256

    Gradient descent is to AI, what a loop is to understand programming. However, understanding a loop doesn't mean you can program a full video game from scratch. Organizing hundreds …

  17. comment
    Comment #44402986

    The problem of measuring intelligence is that most of these techniques reveal more about what we think intelligence is in a specific cultural context. Which means that individuals …

  18. comment
    Comment #44281128

    What I do is to build a substring out of the initial strings that is a multiple of 2. If the string I try to search is 9 characters long, then I extract an 8 characters substrings …

  19. comment
    Comment #44274902

    I implemented many different algorithms for searching and splitting strings using SMID methods as well: https://github.com/naver/tamgu/blob/06aedf2f14895925d7b5a8e2... I used a dif…

  20. comment
    Comment #44266460

    Or a variation of Princess Bride mind challenge...

  21. comment
    Comment #44244957

    Seriously, most fine-tuning now is done with LoRa adapters. They are much faster and more reliable. In my lab, I don't know anybody who is trying to do any kind of thorough fine-tu…

  22. comment
    Comment #43885474

    As Linus Thorval said in an interview recently, humans don't need AI to make bugs.

  23. comment
    Comment #43641451

    If you have a look on the GitHub, the oldest files are 5 years old, way before LLMs...

  24. comment
    Comment #43619069

    As I said many times, I have been in the game for 30 years. I started doing AI with rules as early as the beginning of the 90 and I never...never expected to see anything like LLMs…

  25. comment
    Comment #43543705

    Who needs culture when you can buy guns...