yungtriggz

6 ms·

1. 1.▲
   by yungtriggz 2y ago

   80
2. 2.▲
   by yungtriggz 2y ago

   Thanks! And yes, doing this too
3. 3.▲
   by yungtriggz 2y ago

   Thank you! We use YouTube's iframe api + embeddable player as per their ToS - the rest is just css
4. 4.▲
   by yungtriggz 2y ago

   godspeed
5. 5.▲
   by yungtriggz 2y ago

   Thanks - appreciate the feedback on the contrast. And interesting, I feel like I haven't seen that happen with longer videos much but will look into it, thank you!
6. 6.▲
   by yungtriggz 2y ago

   Thanks! We process the videos in parallel so the processing takes as long as the longest uncached transcript. But then we cache those transcripts so if the videos come up later they don't take as long to process. We're planning on
7. 7.▲
   Google DeepMind has a new way to look inside an AI's "mind"(technologyreview.com)

   4 points by yungtriggz 2y ago|0 comments
8. 8.▲
   How we built a new type of video search

   2 points by yungtriggz 2y ago|2 comments
9. 9.▲
   Show HN: Remy – AI-Curated Video Playlists on Any Topic(useremy.com)

   6 points by yungtriggz 2y ago|6 comments
10. 10.▲
    by yungtriggz 2y ago

    Yea, that tracks. It's wack too because it's affecting the products it's in - like Cursor. It'll make a one-line change in a file but replace every other line in the file with * leave the rest unchanged* . first world pr
11. 11.▲
    by yungtriggz 2y ago

    But it's Cornish
12. 12.▲
    by yungtriggz 2y ago

    Wild that you've posted this once a year for the last 3 years. Love it
13. 13.▲
    by yungtriggz 2y ago

    Yea, at this point I just assume everyone I see online gushing about OpenAI is being paid to do so
14. 14.▲
    Why does everyone keep saying OpenAI's o1 is so much better than Claude?

    2 points by yungtriggz 2y ago|4 comments
15. 15.▲
    Show HN: AI-Curated Video Playlists on Any Topic(useremy.com)

    2 points by yungtriggz 2y ago|0 comments
16. 16.▲
    Show HN: Remy – an AI agent that finds the best clips from the world's videos(getremy.app)

    2 points by yungtriggz 2y ago|0 comments
17. 17.▲
    by yungtriggz 2y ago

    have you considered the fact that they may not want you to have information?
18. 18.▲
    by yungtriggz 2y ago

    yea basically
19. 19.▲
    by yungtriggz 2y ago

    Built something like this a little while ago while iterating on the concept of AI finding relevant video content more generally. I loved it but wasn't maintaining it. Glad to have something I can just use now! Congrats on the launch
20. 20.▲
    by yungtriggz 2y ago

    @forbiddenvoid what is the product you need?
21. 21.▲
    More Meta Layoffs(theverge.com)

    12 points by yungtriggz 2y ago|1 comments
22. 22.▲
    by yungtriggz 2y ago

    agreed, will do. thanks
23. 23.▲
    Show HN: Search engine powered by video clips(searchremy.com)

    4 points by yungtriggz 2y ago|2 comments
24. 24.▲
    Show HN: AI-generated tweets, based on your favourite profiles(mysocialautopilot.com)

    2 points by yungtriggz 2y ago|0 comments
25. 25.▲
    Stanford STORM: An OS research project to explore building AI systems(storm-project.stanford.edu)

    1 points by yungtriggz 2y ago|0 comments
26. 26.▲
    by yungtriggz 2y ago

    This is really cool! I love the rabbit hole stuff you can do when you give GPT more capabilities. I was playing around with this stuff and found I was most often wanting to use it when wanting to learn about something so made Instaclass: h
27. 27.▲
    Why do YOU think Meta and xAI are so desperate to partner with Character.ai?(ft.com)

    1 points by yungtriggz 2y ago|2 comments
28. 28.▲
    by yungtriggz 2y ago

    yea reckon vllm is the way to go. cheers
29. 29.▲
    by yungtriggz 2y ago

    thanks, tried this to some mixed results. seems like they have caps on speed/rate limits etc if you havent spoken to them so might reach out
30. 30.▲
    How fast can one reasonably expect to get inference on a ~70B model?

    9 points by yungtriggz 2y ago|6 comments