ArtRichards

5 ms·

1. 1.▲
   by ArtRichards 16d ago

   I tend to think of it as reappropriating words in a different context. Since we're talking about language models, they're analogues but not as we would assign the same meaning to other humans.
2. 2.▲
   by ArtRichards 21d ago

   I'll drop mine here, I'd really like feedback, though I guess everyone has their own by now, very few will try a new SWE skills framework. https://artrichards.github.io/agent-playbook-suite/blog/ it work
3. 3.▲
   by ArtRichards 24d ago

   I have one I started back in October last year and been improving since. Also a python docs cli to manage agent state in markdown locally. I love reading them to see if theres a way mine could be improved, what topics they hit on, how they
4. 4.▲
   by ArtRichards 1mo ago

   I tried setting up opencode for the server, but it just won't persist my conversations. Hows the mobile/client access on this? Claude remote works well... for Claude... but I havent seen anything comparable open source. edit:
5. 5.▲
   by ArtRichards 1mo ago

   I always think of it like: let it activate whatever experts to reason/do the thing, then in the next prompt give me your explainer experts. Then, both are in the conversation history.
6. 6.▲
   by ArtRichards 1mo ago

   Yes but thanks to the soul or system or whatever (talk at the operator's level), 'layman' feels like it now tries to oversimplify!
7. 7.▲
   by ArtRichards 1mo ago

   Basically every decision or summary I now ask "Help me understand better, please explain in plain-language with examples." Works with 5.6 sol also, when you're deep in the weeds. I rationalize this as the models attempting t
8. 8.▲
   by ArtRichards 2mo ago

   New, smaller models can outperform the previous generation's foundation models. What if there's a way to extract the commodity of intelligence from smaller models? I've seen for many use cases it's well enough. :)
9. 9.▲
   by ArtRichards 2mo ago

   I wonder if this was made in the sub agent's chat? Thanks for sharing.
10. 10.▲
    by ArtRichards 3mo ago

    I saw another post today about a startup making an oven for baking bread. I feel that often the main issue lies with what you need 'the code' to accomplish. My flow state is thinking about and understanding this: am I solving a
11. 11.▲
    by ArtRichards 4mo ago

    They're doing some really cool stuff with Agents too.. automating software pm. it's a huge difference between this and their (unnamed) larger competitors.
12. 12.▲
    by ArtRichards 4mo ago

    Its all about retaining the context and spawning sub agents which can bootstrap quickly and accurately. I'm interested in others dping something similar :) I included a docs cli tool in pypi to manage this context: https://
13. 13.▲
    by ArtRichards 4mo ago

    This is the part I think we will see become more relevant. I created docs-cli (pypi) to manage the index of specs as source code: the framework that goes with it will first create tests for as much as it can, so reproducability becomes the
14. 14.▲
    Show HN: docs-cli - coding-agent project state in Markdown(artrichards.github.io)

    6 points by ArtRichards 4mo ago|0 comments
15. 15.▲
    by ArtRichards 6mo ago

    This article was a really uplifing take... Happy to see more about how awesome we can be, when we care.
16. 16.▲
    by ArtRichards 6mo ago

    Apparently there's been work to expose Meta pushing/funding this, to shift age responsibility from them and force for fine grained age data to be provided to apps.
17. 17.▲
    by ArtRichards 7mo ago

    I like using the automatic lists in soundcloud to discover new music. Often its hit or miss but it can surface some great tracks... Its intentional though, gotta have your finger on the skip track and heart...
18. 18.▲
    by ArtRichards 7mo ago

    This is exactly how I've found leads to most consistent high quality results as well. I don't use gemini yet (except for deep research, where it pulls WAY ahead of either of the other 'grounding' methods) But Codex to p
19. 19.▲
    by ArtRichards 8mo ago

    Sure would love to be able to read this...
20. 20.▲
    by ArtRichards 9mo ago

    I guess he's being a bit cheeky ;)
21. 21.▲
    by ArtRichards 9mo ago

    Is this the same Google that is preventing us from installing unapproved software on our phones?
22. 22.▲
    by ArtRichards 10mo ago

    and > But EasyJet says it has already completed the required software update and is planning on operating its flights as normal on Saturday
23. 23.▲
    by ArtRichards 10mo ago

    Lurking for over 20 years, through my entire professional career... High School to Middle Age... Thanks Dan & HN Community
24. 24.▲
    by ArtRichards 11mo ago

    In a youtube demo, the author says he has been working on this idea for a while, and was motivated to deliver after the GIF. So, kinda? Thanks for the reference. https://youtu.be/v9FMp5QgOvY?si=iKXJSNIuLiU5nOnj
25. 25.▲
    by ArtRichards 1y ago

    came here to post this, but it shares a different clipboard than ctrl+c/v or something?? anyone know the differences?
26. 26.▲
    by ArtRichards 1y ago

    What did this site just attempt to do with my certificate store?
27. 27.▲
    by ArtRichards 1y ago

    I was looking for this comment! That being a possible reason why certain words alleviate, they actually operate at a different level in our conciousness.
28. 28.▲
    by ArtRichards 1y ago

    Thank you for this link.
29. 29.▲
    by ArtRichards 2y ago

    In Cursor I try to mention specifically and only the files where the changes need to be made. Also, I use HomeAssistant for my Dreame vacuum, Hue lights, electricity monitoring, all hooked up with a chat GPT plugin and TTS STT... its the d
30. 30.▲
    by ArtRichards 2y ago

    Reminds me of this: ‘The machine did it coldly’: Israel used AI to identify 37,000 Hamas targets https://www.theguardian.com/world/2024/apr/03/israel-gaza-ai...