Live data from Hacker News

Viewing profile — binocarlos

binocarlos

HN member
Joined
Mon, Feb 18, 2013, 1:19 PM UTC
HN karma
328
Public activity
120 items

About binocarlos

asyncrocity

Recent public activity

  1. comment
    Comment #47597825

    I find Claude Code features fall into 2 categories, "hmmmm that could be actually useful" vs "there is more kool aid where that came from"

  2. comment
    Comment #46367971

    > I mashed F5 like a degenerate. I love the style of this blog-post, you can really tell that Luke has been deep down in the rabbit hole, encountered the Balrog and lived to tell t…

  3. comment
    Comment #46256459

    I use https://github.com/braden-w/whispering with an OpenAI api key. I use a keyboard shortcut to start and stop recording and it will put the transcription into the clipboard so I…

  4. comment
    Comment #43764569

    I agree with this - I hear a lot of hate towards vibe coding but my experience with voice dictation and using 20 years experience in the trenches and so being very specific telling…

  5. comment
    Comment #41669355

    I would pay hundreds of dollars per month for the combination of cursor and claude - I could not get my head around it when my beginner lever colleague said "I just coded this whol…

  6. comment
    Comment #40371140

    I helped work on the RAG part of this :-) We used https://github.com/pgvector/pgvector under the hood and found it extremely easy to integrate with our database schema - being able…

  7. comment
    Comment #39304198

    Excellent link thank you - I will make sure we check this out because as the README says: > Finetune Mistral, Llama 2-5x faster with 70% less memory! Could be very useful for us! D…

  8. comment
    Comment #39304177

    Great question! scheduling workloads onto GPUs in a way where VRAM is being utilised efficiently was quite the challenge. What we found was the IO latency for loading model weights…

  9. comment
    Comment #39304121

    We are finding that fine tuning is very good at setting the style and tone of responses. A potential use case we are thinking about is what if your star sales person leaves the com…

  10. comment
    Comment #39304077

    Thanks for the feedback - I agree that fine tuning a) has potential and b) is not easy :-) > Also, I don’t get the comparison of rag vs finetuning in articles like this - why not d…

  11. comment
    Comment #39302690

    This is spot on, the thing we've not yet done is make it easy to import a repo(s) code and the associated metadata into a fine tuning session easily. > I often wonder how you'd go …

  12. comment
    Comment #39302647

    Fine tuning on your documents will really help to answer questions in the style and tone of those documents, so in that way, yes it helps. It would be possible to include some part…

  13. comment
    Comment #38611002

    I use hacker news as my "is my Internet working" test - it's fast and always up (kudos) So yes, I also thought my Internet was broken :-)

  14. comment
    Comment #35060674

    This is a really cool blogpost - that you can influence a model with guidance towards a desired outcome feels like a really valuable feature. I'm currently working on a project tha…

  15. comment
    Comment #27771297

    thanks! we were stuck with exactly that problem and decided "it's David Bowie" let's try something different :-)

  16. story
  17. comment
    Comment #23143155

    We are thinking for people writing documentation, blogs or Intranets where there are many different pages and there is a more polished sense of branding. You are quite right that i…

  18. comment
    Comment #23140031

    Thanks for the feedback! Yes - dealing with access to a users entire drive is something we are looking at to try and get working for exactly that reason - the words `this app can d…

  19. comment
    Comment #23139752

    We've been working hard on https://nocode.works - which turns Google docs into nice looking websites - it's not pluggable yet but it doesn't require any setup on behalf of the user…

  20. comment
    Comment #23139736

    Shameless plug: We've been working hard on https://nocode.works - which turns Google docs into nice looking websites. We think it's perfect for documentation for non-technical comp…

  21. comment
    Comment #16679957

    Perhaps part of that was that IE was really bad at it's job and as users were told about Chrome and Firefox and they tried it and it was much faster - they switched because IE was …

  22. comment
    Comment #16583611

    Take Basic Attention Token ( https://basicattentiontoken.org/ ) as an example - it is attempting to remove YouTube as an advertising middle-man that controls and takes a cut of ad …

  23. comment
    Comment #16491989

    I use redux a lot and love the way you can reason about the state of the app from a single object - which opens up all sorts of useful things like time-travel etc. However - after …

  24. comment
    Comment #16328475

    Thanks for the feedback - "docker push " is nice :-) For the same reasons as you mention in your post, we've been hard at work with Kubernetes support - Persistent Volumes but with…

  25. comment
    Comment #16328436

    Thats the idea yes! you can run dotmesh on your own servers and install locally on each users machine to then push and pull just like git remotes. It's using copy-on-write so you a…