sonnig

6 ms·

1.
by sonnig 8mo ago
I've owned iOS devices since around 2010 I think, and one of the first things I do when setting them up is disable autocorrect and most other typing assistances. For this reason! Also when typing in 2 or 3 languages, it is completely u
2.
by sonnig 8mo ago
Yep, it's similar in that way, but not in a imageboard/discussion context
3.
by sonnig 8mo ago
Yeah me too
4.
by sonnig 8mo ago
Apologies for that, I have removed it.
5.
by sonnig 8mo ago
I've changed this now, thanks for the feedback
6.
by sonnig 8mo ago
i agree removing punctuation wouldve been a good idea alas it may be a bit too late since that would modify the hash of previous inputs in the future hmm but i will think about it
7.
by sonnig 8mo ago
Me too, and other 16 users
8.
by sonnig 8mo ago
True! That would be a more powerful approach. Here I kept it quite basic since I was not very familiar with the tooling. I do apply lowercasing of text + some whitespace stripping in order to increase the number of collisions a bit. Edit:
9.
Show HN: Subth.ink – write something and see how many others wrote the same(subth.ink)
92 points by sonnig 8mo ago|51 comments
10.
by sonnig 8mo ago
Assuming both agents are using the same model, what could the reviewer agent add of value to the agent writing the code? It feels like "thinking mode" but with extra steps, and more chance of getting them stuck in a loop trying to
11.
by sonnig 8mo ago
Well put. I can't help thinking of this every time I see the 854594th "agent coordination framework" in GitHub. They all look strangely similar, are obviously themselves vibe-coded, and make no real effort to present any type
12.
by sonnig 8mo ago
I am working with kilo-stock-tokens. Currently producing 3000 LoC/h (trying to ramp up to 6000 by the end of the week). I have also deployed 4 union-busting agents in case the other agents decide to quit all at once.
13.
by sonnig 8mo ago
I searched for "poke push notification framework" on Kagi. But even then it was hard to find.
14.
by sonnig 8mo ago
Mind sharing your workflow? I'm at 24.3x productivity right now, 5 parallel agents, 2 monitoring Opus agents, 1 architect agent and 2 Senior QA agents, each with independent memory and 12 MCP servers. They are running in 78 parallel ta
15.
by sonnig 8mo ago
It makes me so exhausted trying to read them... my brain can tell immediately when there's so much redundant information that it just starts shutting itself off.
16.
by sonnig 9mo ago
Maybe [1]? Sounds like ntfy.sh, but more… AI [1] https://poke.com/docs/developers/api/message-poke
17.
by sonnig 9mo ago
I think the same. It sounds quite more practical to have LLMs code in languages whose compilers provide as much compile-time guardrails as possible (Rust, Haskell?). Ironically in some ways this applies to humans writing code as well, but t
18.
by sonnig 11mo ago
I strongly recommend the book "Mastering Emacs" by Mickey P. You will need some patience as well, I recommend going slow and steady for a week at least using the book, with a vanilla/standard Emacs install. It took me about 2
19.
by sonnig 11mo ago
This conversation reminds me a bit of https://news.ycombinator.com/item?id=21191676 some time ago (similar misuse of the term "open source")
20.
by sonnig 11mo ago
Looks great!
21.
by sonnig 11mo ago
we've integrated granian into our stack at $WORK recently and couldn't be happier
22.
by sonnig 1y ago
Are many HN members in Zürich? I was suprised to see this type of news with so many points, though it's quite amusing to see the mixture of different topics in the front page.