ai_slop_hater

7 ms·

1. 91.▲
   by ai_slop_hater 6mo ago

   I did consider some factors, including the ones you mentioned, when drafting my original message. I just didn't include my considerations into my message. I prefer addressing factors on-demand rather than attempting to immediately addr
2. 92.▲
   by ai_slop_hater 6mo ago

   > (Edit to add: this is just a comment on Kubernetes being invoked whenever someone talks about scalability; I have massive respect for what the TigerBeetle folks are doing) Me too. Why did you have to add this edit though? Is there anyt
3. 93.▲
   by ai_slop_hater 6mo ago

   If you think that my opinion is not worth listening to, i.e. that I am wrong, would you mind elaborating why? There is a real opportunity to sway my opinion here, because I am not unsure. I could just be crazy. I don't know anymore. Bu
4. 94.▲
   by ai_slop_hater 6mo ago

   > But, why? To keep things simple. My current company is running multiple instances of back-end services for absolutely no fucking reason, and I had to fix numerous race condition bugs for them. I had an interview with a startup where, a
5. 95.▲
   by ai_slop_hater 6mo ago

   The takeaway is that you can just buy a single beefy server instead of using kubernetes or whatever.
6. 96.▲
   by ai_slop_hater 6mo ago

   This matches my experience as well, "adaptive thinking" chooses to not think when it should.
7. 97.▲
   by ai_slop_hater 6mo ago

   You can try implementing rate limiting and not exposing your API keys to the public.
8. 98.▲
   by ai_slop_hater 6mo ago

   We no longer need dashboards, just connect ChatGPT to a metrics database or whatever.
9. 99.▲
   by ai_slop_hater 6mo ago

   https://news.ycombinator.com/item?id=47779730
10. 100.▲
    by ai_slop_hater 6mo ago

    Unfortunately OpenAI's models really suck at JJ, while, on the other hand, Opus 4.6 is much better at it, probably due to a later training cut off.
11. 101.▲
    by ai_slop_hater 6mo ago

    Because with RSC you don't have a shitload of loading indicators and layout shifts.
12. 102.▲
    by ai_slop_hater 6mo ago

    https://github.com/cloudflare/serverless-registry
13. 103.▲
    by ai_slop_hater 6mo ago

    It is absolutely bizarre to not check DevX for Rust, it was literally the most loved programming language for years according to SO surveys.
14. 104.▲
    by ai_slop_hater 6mo ago

    > Most of the Rust code I've read is arc-mutex-slop. Skill issue
15. 105.▲
    by ai_slop_hater 6mo ago

    Can you share which difficult and non-intuitive corner cases there are? I guess debouncing, etc.
16. 106.▲
    by ai_slop_hater 6mo ago

    It's so funny how all these billion dollar AI companies seem to be unable to hire a single person who can code.
17. 107.▲
    by ai_slop_hater 7mo ago

    Then it's a stupid law. But from the image that other commenter gave, it does look like Germany has space that is clearly intended for cyclists, and I have no issue with that. I have issue with instances when people cycle on sidewalks
18. 108.▲
    by ai_slop_hater 7mo ago

    I am aware that most countries do not have dedicated roads for cyclists, but that doesn't mean that cyclists should be using sidewalks. When I go out and walk on the sidewalk, I expect to be able to just walk safely without having to t
19. 109.▲
    by ai_slop_hater 7mo ago

    I think they just want LLMs to read the docs they began shipping[0] along with the library instead of using their own knowledge. For example, when I used Next.js a few months ago, models kept using cookies() and headers() without await, bec
20. 110.▲
    by ai_slop_hater 7mo ago

    [flagged]
21. 111.▲
    by ai_slop_hater 7mo ago

    When you create a Next.js project from Vercel's template, you get an AGENTS.md that literally says "THIS IS NOT THE NEXT.JS YOU KNOW"
22. 112.▲
    by ai_slop_hater 7mo ago

    Maybe, but you can literally feel the difference as you type. When you type in Codex, it's fast, it feels instant. When you type in Claude Code, it feels like playing a game in 60 fps after you already got used to 144 fps.
23. 113.▲
    by ai_slop_hater 7mo ago

    I didn't use commands. I only used rules, memories, and skills. I asked Codex to read rules and memories from where Claude Code stores them on the filesystem and merge them into `AGENTS.md` and this actually works better because Anthr
24. 114.▲
    by ai_slop_hater 7mo ago

    Alright, I just tested that setting and it doesn't work. https://i.imgur.com/MYsDSOV.png I tested because I was porting memories from Claude Code to Codex, so I might as well test. I obviously still have subscription d
25. 115.▲
    by ai_slop_hater 7mo ago

    Does that mean we should use a larger model as judge for evals, not a smaller one?
26. 116.▲
    by ai_slop_hater 7mo ago

    I read his comment and still replied. I think his claim that nobody reads thinking blocks and that thinking blocks increase latency is nonsense. I am not going to figure out which settings I need to enable because after reading this thread
27. 117.▲
    by ai_slop_hater 7mo ago

    Thanks. I was referring to the fact that Anthropic, in their API, prohibits setting temperature when thinking is enabled.
28. 118.▲
    by ai_slop_hater 7mo ago

    Huh, I wonder if that's why you cannot change the temperature when thinking is enabled. Do you have a link for the paper?
29. 119.▲
    by ai_slop_hater 7mo ago

    > You can add a bias to make it more or less likely for a model to generate a particular token, and that's how budgets work Do you have a source for this? I am interested in learning more about how this works.
30. 120.▲
    by ai_slop_hater 7mo ago

    I also have similar experience with their API, i.e. some requests get stalled for minutes with zero events coming in from Anthropic. Presumably the model does this "extended thinking" but no way to see that. I treat these requests