unshavedyak

5 ms·

31.
by unshavedyak 4mo ago
I wouldn’t let anyone else use it in general because it’s an asset i don’t want to lose. As it is I now have to be careful what I use it for as I don’t want to trip any flags. Eg today I was curious how some LLM benchmarks worked and I want
32.
by unshavedyak 4mo ago
Yea but i don't think the LLM can inspect that. Eg it's similar to a code function , the code executes but the code doesn't have the ability to inquire about the code itself. In that silly example/thought, the LLM would
33.
by unshavedyak 4mo ago
I often wonder how humans "know" things. I suspect (ignorant armchair) we have some ability to signal strength of those facts, via repetition. Without this layer of introspection i imagine LLMs can never avoid hallucination. It ob
34.
by unshavedyak 4mo ago
Yea it's definitely portable, it's just not a friendly formfactor for where my compute sits, where my keyboard sits, etc. If i'm in a chair at the part i'd need a literal lap-top, three components (keyboard, compute, mon
35.
by unshavedyak 4mo ago
I'd buy it but i want it in a laptop form or maybe tablet, or something. Being a monitor means the usefulness for me, ie being able to program outside, is kinda moot.
36.
by unshavedyak 4mo ago
> There is no trusted middleman who does the verification and takes the burden of accountability. It's a great example of a project that needs accountability, but there's also thousands of other projects that need absolutely no
37.
by unshavedyak 4mo ago
Depends on the scope. I’m slopping some projects (graphics related) that I’d never have learned before. Mostly because of the huge lift, huge time investment, etc. is it quality code? I’m sure not. But verification of the application is qui
38.
by unshavedyak 4mo ago
Would be nice if they’d add in a simple back off retry mechanism.
39.
by unshavedyak 4mo ago
which is hilarious because i was excited when i heard Digg was coming back. Many platforms are having a difficult time with bots, mass thread manipulation, etc. I'd be interested in a platform which attempted to fix that problem. I tho
40.
by unshavedyak 4mo ago
> While Americans Oppose AI Data Centers I know it's more mixed and complex than this, but i think a big opposition is not to the data centers themselves but to their locations. Too often it feels like the centers are exploiting loc
41.
by unshavedyak 4mo ago
This is very close to what i've been wanting. I'm slopping' an IDE to eventually explore this space as well. My general thought is AI should be monitoring what i write for code, and not only make suggestions, but allow using
42.
by unshavedyak 4mo ago
> And if you used it to poke around, you would also agree. Would you elaborate? Not sure what you're describing
43.
by unshavedyak 4mo ago
Nope, i'm on x20 and almost exclusively use Claude Code. I have a pretty bare bone setup with some custom hooks, skills, etc. I try to keep context lean so i don't like to add much stuff.
44.
by unshavedyak 4mo ago
It's funny, i'm getting close to not caring anymore how much better a model is. I want it to be about as good as 4.8, but most importantly to be very good at following directions, style, etc. I really like Claude for that in gener
45.
by unshavedyak 4mo ago
Man did I despise that keyboard. I went from hating that to adoring my M3. Which feels good because I loved my MacBook before those butterfly switches, and again I love my MacBook now. I almost left apple entirely over those stupid switches
46.
by unshavedyak 4mo ago
> not open the music player? I'd be fine with it doing that if it actually opened what i listen with. The OS can clearly see i spend 100% of my time in another music player (Spotify), opening Apple Music is at best a poorly designed
47.
by unshavedyak 4mo ago
> Discussions about choosing a library with the best syntactic sugar method naming is just as crazy as suggesting we type in assembly. I have a more hopeful take. As AIs improve and get faster we can more quickly and iteratively improve
48.
by unshavedyak 5mo ago
> Am I Unc? > Take the Unc Test and find out how cooked your aura really is. Okay, as i get older i run into more of these scenarios where the new lingo is mostly beyond me. However this is not a complaint, if anything i'm more c
49.
by unshavedyak 5mo ago
Not the parent, but i assume these are relevant: - https://web.archive.org/web/20260602130637/https://www.techn... - https://web.archive.org/web/20260520190620/https:/&#x2F
50.
by unshavedyak 5mo ago
Are they mutually exclusive?
51.
by unshavedyak 5mo ago
Not the parent, but someone else commented on this: https://news.ycombinator.com/item?id=48347735
52.
by unshavedyak 5mo ago
What sort of differences beyond gym related?
53.
by unshavedyak 5mo ago
I really wish it was more common to use AI for augmenting than authoring. Eg i find coding with LLMs neat when you primarily "talk" to it through code, by filling out structs, funcs, fields, etc - where it would use your changes a
54.
by unshavedyak 5mo ago
> Edit: i bet 99% of people here, if presented with a test where i gave 5 models but all of the results came from one, would not be able to discern this. Just vibes all the way down. I think you're missing one (or more) of the facet
55.
by unshavedyak 5mo ago
> Subscription-capable - Anthropic Claude Pro/Max (anthropic), OpenAI Codex / ChatGPT Plus/Pro (openai-codex), Kimi Code (kimi), and GitHub Copilot (github-copilot). Am i reading this right? Seems to suggest that this can
56.
by unshavedyak 5mo ago
The ACP budget change is so bizarre to me. If i was more adventurous with my subscription i'd be interested to see if you could intercept UI/input from CC TUI and render that in a native GUI without it being a TUI. That would be &
57.
by unshavedyak 5mo ago
Oh, yea here's all the proof you need. Even Zed themselves admit you won't be able to use Claude Code via ACP via Subscription: https://zed.dev/blog/terminal-threads So yea, bcherny didn't reply to me bu
58.
by unshavedyak 5mo ago
Which seems logical - if they were somehow consistent between all LLMs i'd be even more curious how LLMs are... crab-evolving to 'isms, but hopefully that's not the case hah. (i say crab-evolving in jest)
59.
by unshavedyak 5mo ago
to be clear, i'm referring to the recent fact where it appears that they're disabling all Claude Code (Subscription) usage from the SDK. Which ACP would be included on. As usual though it's not super clear exactly what is all
60.
by unshavedyak 5mo ago
VSCode has an official client? Given IDE usage is being restricted from Claude Code via the CC SDK tokens going to the Claude API rather than your CC Subscription, i'm unclear which IDEs can actually use claude code now. Eg is Zed capa

More ›