preyneyv

3 ms·

1.
We're Learning Backwards: LLMs build intelligence in reverse(pleasedontcite.me)
3 points by preyneyv 6mo ago|1 comments
2.
by preyneyv 6mo ago
Wrote this after a lecture on Hays & Efros' scene completion using just lookup on 2.3M images. Made me think about where the "simple model + lots of data" bet actually leads. The article traces that thread through the Sca
3.
Show HN: I turned ARC-AGI-3 into a daily browser game(arcaptcha.io)
4 points by preyneyv 6mo ago|0 comments
4.
by preyneyv 2y ago
ooh these are excellent, thank you for sharing!
5.
by preyneyv 2y ago
Yeah, it's definitely one of those things where your eyes are "opened" when you use something faster. Kind of like headphones, what you have is good enough until you try something better.
6.
by preyneyv 2y ago
Hard agree on VR / AR latency requirements. It's the first consumer use case in a while that's ahead of the average user's compute power, and I really hope it will mean good things for performant software.
7.
by preyneyv 2y ago
I think you have it backwards. We get away with longer distances because of faster automobiles. But imagine for a moment what would happen if we had faster automobiles and shorter distances: things would just be better.
8.
by preyneyv 2y ago
This is exactly the issue. We can do things faster => we can do more things worse!
9.
by preyneyv 2y ago
Honestly, this was a big one for me too. I remember being a kid and using Office 2007 on a terrible Windows XP desktop, and things just worked. Now I'm on a "modern" machine with "modern" software, and right-click t
10.
I Waited 10B Cycles and All I Got Was This Loading Screen(blog.preyneyv.dev)
65 points by preyneyv 2y ago|59 comments
11.
by preyneyv 2y ago
complete satire really fun to build though
12.
Show HN: Use prompt": Compile-time code generation(twitter.com)
1 points by preyneyv 2y ago|2 comments
13.
HTTP/1.1 500 Segfault: Building a web server in C(blog.preyneyv.dev)
1 points by preyneyv 2y ago|0 comments