esafak

15 ms·

451.
by esafak 4mo ago
Surely the AI champion can share a script to keep everyone's skills up to date?
452.
by esafak 4mo ago
Could you make the tables sortable?
453.
by esafak 4mo ago
It is the same with anything technical; you need taste and wisdom, which are borne of experience. I think society will want to subsidize this learning lest it deskills en masse.
454.
by esafak 4mo ago
I think so. Mathematics is the least of my worries. I worry about what the machines will want to do.
455.
by esafak 4mo ago
> What's left? I think humans will be left to propose new conjectures while machines fill out the proofs. I don't know if there are enough interesting conjectures to go round to build new careers, though.
456.
by esafak 4mo ago
On another OS tooling note, jdx is joining entire.io while continuing to work on mise: https://jdx.dev/sponsors.html
457.
by esafak 4mo ago
Just run weekly cron job to assess code quality and highlight candidates for refactoring. In addition to doing the same in each PR, of course, but things can get through.
458.
by esafak 4mo ago
I do that with my daughter and my wife says I'm teaching her to hate everything, emanating bad energy. It is tough to have higher standards than people around you. They're happy, you're the downer.
459.
by esafak 4mo ago
We found the prompt, guys!
460.
by esafak 4mo ago
Prose that reads as artificial as the 'dairy dessert' it describes.
461.
by esafak 4mo ago
I think it does not use JEPA.
462.
by esafak 4mo ago
That is not how it works. IF it was condemned to be average models wouldn't be constantly improving, given that humans aren't getting better.
463.
by esafak 4mo ago
An economics article that completely ignores the cost of bugs!
464.
by esafak 4mo ago
Just use the web/desktop app and you won't need to.
465.
by esafak 4mo ago
The DIA's Maven database was out of date: https://www.theguardian.com/news/2026/mar/26/ai-got-the-blam...
466.
by esafak 4mo ago
Does it support subagents?
467.
by esafak 4mo ago
ASI is going to be here by the time Lecun gets started.
468.
by esafak 4mo ago
The hardest problem is social. Who is going to use this?
469.
by esafak 4mo ago
Which is funny because OpenAI talks up the value of their ChatGPT brand. Only coders know Codex.
470.
by esafak 4mo ago
Anybody using this in production? edit: Thanks, will try!
471.
by esafak 4mo ago
I hope that's not the future of work before UBI gets settled, because almost nobody grows their own food.
472.
by esafak 4mo ago
How do you like its design mode?
473.
by esafak 4mo ago
He's just got a case of the Mondays.
474.
by esafak 4mo ago
Those languages dominated because they were simple. Then they grew, and their users grew up, and realized that worse is better. At a startup you can choose even fancier languages, since nobody is stopping you!
475.
by esafak 4mo ago
Ugh, that changes everything. If I wanted an arranged marriage I could go back to Claude Code.
476.
by esafak 4mo ago
That's an ask for Apple, not us ;)
477.
by esafak 4mo ago
It says Stan and PyMC beat Noise at the thing they’re built for, fitting a posterior to lots of continuous data with their HMC/NUTS samplers, and NumPy beats it at raw array crunching. Conditioning in Noise is rejection-based, so it w
478.
by esafak 4mo ago
Skills provide guidance; they augment and narrow the search space. Even intelligent humans benefit from guidance.
479.
by esafak 4mo ago
https://en.wikipedia.org/wiki/Don't_Copy_That_Floppy
480.
by esafak 4mo ago
Somebody needs to be responsible for quality and say 'no' if not the CEO.

More ›