merlindru

7 ms·

61.
by merlindru 5mo ago
Applied!
62.
by merlindru 6mo ago
SEEKING WORK | Germany | Remote Technologies: Anything. Very strong with TypeScript, frontend (React, Svelte, Vue, etc.) and backend. I love Bun. Very strong with Rust. Have used many languages. I've built getinvoke.com - an app that l
63.
by merlindru 6mo ago
but that doesn't have anything to do with LLMs. if someone made the same gigantic mess of a PR without LLMs, it would still be rejected, because it is a gigantic mess of a PR. the low effort part is the problem. what if i made a great,
64.
by merlindru 6mo ago
Also huge opportunity for scams etc if this ever was a targeted takeover type thing. Emails and other stuff go to the same domain, and an impostor could just keep answering correspondence like nothing had happened And even worse, if I wante
65.
by merlindru 6mo ago
but why? why would an insider put the wrong domain into a strangers account that has no interest in using the domain and went out of her way to give it back to the rightful owners?
66.
by merlindru 6mo ago
highly agree, sadly, as a huge fan of Opus Opus 4.5 and 4.6 were the first models that i could talk to and get a sense that they really "understood" WHY i'm saying the things i am Opus 4.7 kinda took that away, it's a de
67.
by merlindru 6mo ago
Anthropic has started to ask for IDs for use of their products period I don't like that trend. I get why they're doing it, but I don't like it
68.
by merlindru 6mo ago
I like VSCode and can't really switch to Zed, but Zed has two very good autocomplete models their own, called Zeta 2 and then Mercury by Inception. May also be available in VSCode through some third party extension like Kilo, not sure
69.
by merlindru 6mo ago
that's akin to saying "i do not need their product therefore i don't care"... so what's your point? someone may have made it part of their workflow!
70.
by merlindru 6mo ago
about once a month since my childhood i get an awful, slowly ramping headache that makes it unable for me to think or function. it seems to happen more when i'm overweight, making me think it's blood pressure (BP) related, but the
71.
by merlindru 6mo ago
dunning krueger in the training materia
72.
by merlindru 6mo ago
i've noticed this too. i suspect they either did weight surgery on the model and distilled it from Mythos Preview, or they are currently not saving it correctly / having another adaptive thinking bug. i asked it to look at writing
73.
by merlindru 6mo ago
They swapped the tokenizer which either means a new pretrain, or token/weights surgery. The latter one seems more likely both because - economics: i'd wager a bet that Opus 4.7 is just distilled Mythos Preview - performance: surge
74.
by merlindru 6mo ago
one thing to keep in mind is that you have to use GPT-5.4 differently from codex. they "work" in different ways. i was aghast when i noticed how terrible Codex was against Claude Code only to conclude it was me who wasn't usi
75.
by merlindru 6mo ago
but how true is this? this is almost impossible to measure and those that do[1] find no significant difference i personally haven't noticed any downgrade at all. it's entirely possible there's a mass delusion going on where e
76.
by merlindru 6mo ago
Yes but only for the message that includes it. Whereas /effort max keeps it at max effort the entire convo, to my knowledge
77.
by merlindru 6mo ago
it may be the agent features in my case. now that i think about it, i also forgot that my CLAUDE.md is different from my AGENTS.md either way, all that one can really rely on is the benchmarks, and those are easily cheated/overfitted t
78.
by merlindru 6mo ago
i think it may matter a good bit. i definitely have to write in different styles with different models (and catch myself doing so unintentionally) now that you mention it... definitely not bologna, at least anecdotally :)
79.
by merlindru 6mo ago
very interesting. i think "verbal biasing" and "knowing how to speak" in general is a really important thing with LLMs. it seems to massively affect output. (interestingly, somewhat less with Opus than with GPT-5.4 and C
80.
by merlindru 6mo ago
yes but i can identify those easily. i know that if it flags something that is obviously a non issue, i can discard it. ...because false positives are good errors. false negatives is what i'm worried about. i feel massively more sure t
81.
by merlindru 6mo ago
I like biasing it towards the fact that there is a bug, so it can't just say "no bugs! all good!" without looking into it very hard. Usually I ask something like this: "This code has a bug. Can you find it?" Sometim
82.
by merlindru 6mo ago
i tried out gpt 5.4 xhigh and it did meaningfully worse with the same prompt as opus 4.6. like, obvious mistakes
83.
by merlindru 6mo ago
I still haven't got this email, anyone else?
84.
by merlindru 7mo ago
Not a real solution but you could try using AquaVoice for dictation. It can gather screen context so you just say the function name out loud and it capitalizes and spells everything correctly. (Even hard cases!)
85.
by merlindru 7mo ago
Try Mercury by Inception. It's available as autocomplete in Zed. Last time I tried it, Zed had an API key hidden in their docs that allowed you to use it for free The crazy thing is that it's a diffusion-based LLM. That makes it v
86.
by merlindru 7mo ago
why should only profits matter? if i had a killer product today that i just need to sell tomorrow, wouldn't you still invest today knowing i'll probably only start to make money tomorrow (or perhaps next week)? the expectation is
87.
by merlindru 7mo ago
yeah they're moving into the wrong direction as well. not to mention that notarisation is for after-the-fact anyways. malware still slips through (historically true!). it's just supposed to shrink the blast radius AFTER apple know
88.
by merlindru 7mo ago
this is gonna sound ranty, but it's straight from the heart: i think most of them are pointless. not every feature needs to be a new keyword. stuff could be expressed within the language. if the language is so inflexible in that regard
89.
by merlindru 7mo ago
besides it being almost impossible to understand what "the right way of doing stuff" is with Swift (or any bloated language), i absolutely _do_ have to use the keywords. reading someone else's code is part of working with the
90.
by merlindru 7mo ago
this is a bad faith take. i think the website is really cool and doesn't reek of slop at all. what makes you think differently?

More ›