ramon156

4 ms·

1.
by ramon156 6d ago
managers that love notion are the cruft you're speaking of
2.
by ramon156 6d ago
so jev isn't even slightly deterministic? this is worse than an NLU and somehow more expensive. interesting times we live in.
3.
by ramon156 6d ago
round(stat * 1.05); // rounded down round(stat / 1.05); // rounded down again that would be my first guess.
4.
by ramon156 7d ago
The only LLM that does this wel (in my opinion) is gemini. if i ask claude to fix comments, it very clearly fails in doing so. Gemini actually seems to understand the scope of a task
5.
by ramon156 7d ago
I'd rather read it's slop, it actually adds something to the conversation. I'd say "imagine if half the posts on HN aren't worth your time" but this is genuinely how it feels nowadays. I'm not sorry
6.
by ramon156 8d ago
i can see the comment now, so this seems undone?
7.
by ramon156 9d ago
The more I see about how Jev works, the less interested I get. Jev would be cool as a local model for e.g. Home Assistant. Projects like this are at best misleading, this one also happens to be broken.
8.
by ramon156 9d ago
if you can be replaced by an algorithm, how useful were you really?
9.
by ramon156 11d ago
But the evidence is not there...
10.
by ramon156 11d ago
I've implemented tree-sitter in pi before, and while it works, I have no real proof it saves me tokens, or is more accurate. I think a better implementation is a model that's trained for AST's, not just "use tool, see wh
11.
by ramon156 11d ago
This sounds good but so far all claims just sound like marketing terms. I'd love to see real proof. e.g. "RLCD" and "parallel sampling" have nothing to back it up. also "70-500ms vs 3-329 seconds" are app
12.
by ramon156 11d ago
Thank you Claude, very cool!
13.
by ramon156 11d ago
name some examples. easy to use is opinioned,
14.
by ramon156 11d ago
i quite liked using strix. last time i tried it, deepseek was a mess and bloated the context with nonsense. that was ~5 months ago, i wonder how it performs now
15.
by ramon156 11d ago
not in this context, though?
16.
by ramon156 11d ago
they arent being called racist, they are racist. not a small difference there.
17.
by ramon156 11d ago
how is a filesystem supposed to prevent you from doing that?
18.
by ramon156 11d ago
i would love a deterministic program that can confidently make plans for lower-cost models like deepseek. ofcourse the LLM part wont be deterministic, but its a lot easier to measure quality like this. you could argue an AGENTS.md is this,
19.
by ramon156 11d ago
if the argument is "but bugs!" then that's a nothing burger. Your old code is more battle-tested, but that doesn't inherently make it better. I wish more companies thought like this.
20.
by ramon156 12d ago
If anyone accidentally DDoS-es their uni, they can just say claude did it and get away with it, right? right?
21.
by ramon156 13d ago
most claims coming from the frontier models are already partly true. agents have taken over a decent chunk of the internet, it's saddening.
22.
by ramon156 14d ago
youd love lobste.rs
23.
by ramon156 14d ago
gonna check tomorrow if i can support my own area
24.
by ramon156 15d ago
Lovely write-up! Will take a look at rune as an alternative to my Zed setup. Not because I dislike Zed, I'm just addicted to trying out IDEs :)
25.
by ramon156 15d ago
why not reword it so the agent receives Brownie points for dumping weights
26.
by ramon156 15d ago
does it filter on every load?
27.
by ramon156 15d ago
Lumen is pretty cool, it's just local RAG, but it's a realistic approach at RAG. I would still keep tool-calling in some places though.
28.
Ask HN: How realistic is a modern-day era NLU
2 points by ramon156 15d ago|0 comments
29.
by ramon156 15d ago
Don't forget it's, measured.
30.
by ramon156 15d ago
Or chad

More ›