siquick

7 ms·

1. 61.▲
   by siquick 2y ago

   When deepfake images of MicolashKyoka start clogging up image searches you may wish for this feature.
2. 62.▲
   by siquick 2y ago

   I’ve lived in Australia for nearly 15 years and still say “swimming costume” - the locals ridicule me and call me for posh for not saying “bathers”.
3. 63.▲
   by siquick 2y ago

   This is great, thanks.
4. 64.▲
   by siquick 2y ago

   I try and avoid screens before sleep but also use a Kindle for reading before sleeping - wondering if the e-ink displays count as screen time in studies like this?
5. 65.▲
   by siquick 2y ago

   Staring at your phone > unconsciously staring at an increasingly uncomfortable person trying to enjoy a quiet cup of coffee
6. 66.▲
   by siquick 2y ago

   I don’t know a single person who is interested in anything like this, including massive gadgets nerds. Who’s the target market? The Verge reporters?
7. 67.▲
   by siquick 2y ago

   I’m a customer of Unisuper - the daily update emails definitely pinned the blame on the service provider.
8. 68.▲
   by siquick 2y ago

   This is a pretty accurate representation of everyday city life in 2024.
9. 69.▲
   by siquick 2y ago

   What’s the name of this pen? Sounds incredibly useful
10. 70.▲
    by siquick 2y ago

    And to think I was mildly concerned when our app inflated out to 36.8MB recently
11. 71.▲
    by siquick 2y ago

    > Likewise, everybody and their dog has ChatGPT installed It’s a fraction of a percent of the population that have used it.
12. 72.▲
    by siquick 2y ago

    Here you go https://bandcamp.com/
13. 73.▲
    by siquick 2y ago

    I’ve been producing music on both software and hardware for 20 years. Borrowed the EP-133 KO II from a mate and found it to be highly initiative and pure style over substance. I don’t get what its selling point is. OP-1 I can understand but
14. 74.▲
    Dissecting Partial Pre Rendering(nikhilsnayak.dev)

    2 points by siquick 2y ago|0 comments
15. 75.▲
    by siquick 2y ago

    Llamaparse by LlamaIndex is probably SOTA at the moment and seems to have no problems with tables. Pricing is good at the moment too. https://www.llamaindex.ai/enterprise
16. 76.▲
    by siquick 2y ago

    I can’t imagine any serious RAG application is not doing this - adding a contextual title, summary, keywords, and questions to the metadata of each chunk is a pretty low effort/high return implementation.
17. 77.▲
    by siquick 2y ago

    I wonder if people get spooked by the simplicity and think it’s a trick question.
18. 78.▲
    by siquick 2y ago

    This is the reality that most engineers that spend their lives online don’t seem to get. We don’t need you to be the best engineer using the most cutting edge tools that a YouTuber told you about. We need you to be a good colleague who peop
19. 79.▲
    by siquick 2y ago

    I didn’t edit my comment.
20. 80.▲
    by siquick 2y ago

    Thankfully most companies in the world aren’t American “at will” employers and don’t have tens of thousands of employees.
21. 81.▲
    by siquick 2y ago

    Uploading a a photo/file to a server does not a product make yet you referenced Dropbox and Instagram and that’s what they started out as. The UX of the AI applications is the moat and the infrastructure providers behind those applicat
22. 82.▲
    by siquick 2y ago

    > We need a Dropbox or a Slack or an Instagram: something people love that makes their life easier or better. People seem to not mind ChatGPT or Claude and safe to say that a very large majority of AI products are using one of the APIs o
23. 83.▲
    by siquick 2y ago

    This is my experience too. Pedestrian crossings are completely ignored and indicating is optional. Combined with this god awful SUVs, Australia is an unpleasant place to be a pedestrian. Sydney has this weird thing where cars drive full spe
24. 84.▲
    by siquick 2y ago

    A tiny tiny percentage of Instagram users would be considered artists.
25. 85.▲
    by siquick 2y ago

    When you’re creating your embedding you can store keywords from the content (using an LLM) in the metadata of each chunk which would positively increase the relevancy of results turned from the retrieval. LlamaIndex does this out of the box
26. 86.▲
    by siquick 2y ago

    Very cool, going to try on some of our workflows this week. Would love to see web access and RAG (LlamaIndex) integration. Are they on the roadmap?
27. 87.▲
    by siquick 2y ago

    They only finished migrating to Google Cloud last year. https://www.googlecloudpresscorner.com/2023-06-05-UniSuper-A...
28. 88.▲
    by siquick 2y ago

    > I don’t know if it’s (still) in service, but the USA had a rocket that would fly over the rocket fields and send the launch codes and fire the icbms. The fear was, that the soldiers were incapacitated and could not fire them itself. Th
29. 89.▲
    by siquick 3y ago

    What does Airbnb use ML for?
30. 90.▲
    by siquick 3y ago

    Of course. My point is that the app was “complete” a long time along yet they add more and more unnecessary features.