ralusek

6 ms·

1.
by ralusek 6d ago
I also signed up for a new account and it's right back to working how it used to. They absolutely do not consume tokens equally across accounts. I did TONS of work on the new account and barely made a dent, even on Astra. Old account c
2.
by ralusek 7d ago
Apple silicon was not an autopilot decision, kind of an insanely cool decision.
3.
by ralusek 11d ago
Over the course of about a month, this went from being vaguely detectable on my mom to very overpowering (to me). I have a particularly sensitive nose, but now I can be across the room and if she comes over to our house I can smell her by t
4.
by ralusek 13d ago
We know they'll have replaced horses when a car costs 200 horses.
5.
by ralusek 14d ago
Google, famous for not winding things down.
6.
by ralusek 14d ago
Also supports Gaussian Splats for spots: https://skatanica.com/spots/5eyevxzo35tcou1n6doxkw2lx?tab=ex... https://skatanica.com/spots/crelhv2mjf6by8yfeitv74tbo?tab=ex... WASD on desktop, two finger
7.
Show HN: Skatanica – Browse and share skate clips by location(skatanica.com)
13 points by ralusek 14d ago|3 comments
8.
by ralusek 18d ago
> Cache hit % I thought you had to actively manage caches, do you not?
9.
by ralusek 18d ago
I didn't mean people aren't using OpenAI, I just mean not through Cursor. Using OpenAI through codex (cli) is the way you should use it.
10.
by ralusek 19d ago
Cursor is like one of the least economical ways to use Claude and OpenAI, virtually nobody does it. You use Cursor for their autocomplete/composer + grok. For using Claude and Codex, you get immeasurably more bang for your buck just us
11.
by ralusek 19d ago
It's basically not going to affect them at all. I don't know a single person that uses OpenAI models via Cursor. Saw another comment say it was 5%.
12.
by ralusek 19d ago
I just finished making a social application (for skateboarding) with the express intent of making it fully featured/pleasant to use on mobile simply as web app, without login or installation. Web is such a powerful platform and so many
13.
by ralusek 27d ago
Things like headless browsers are really annoying to include in FaaS environments, but are super useful. If they're in the runtime binary already, that'd be great.
14.
by ralusek 1mo ago
One of the most interesting talks on the math behind procedural generation: https://www.youtube.com/watch?v=C9RyEiEzMiU That presentation was foundational for me personally wrt quite a few things.
15.
by ralusek 1mo ago
We're just going to slowly deconstruct every element that could be a factor of intelligence. It's not out-thinking, it's just out-remembering It's not out-thinking, it's just out-working It's not out-thinking,
16.
by ralusek 1mo ago
Codex 5.6 sol is arguably superior to Claude, albeit very close. They're functionally indistinguishable to me, but if you're concerned about $$, Codex gives you much, much more bang for your buck.
17.
by ralusek 1mo ago
Can someone help me understand the deep fake controversy? That's like making photoshop illegal.
18.
by ralusek 1mo ago
Mental, that
19.
by ralusek 1mo ago
Is that what the identity function is?
20.
by ralusek 1mo ago
Can maybe think of one time in the last 10 years that I would have wanted this. Its only purpose is for training LLMs
21.
by ralusek 1mo ago
If you're ##CCFFFF please go to the hospital
22.
by ralusek 2mo ago
Socialism is completely incompatible with free markets. Free markets typically stop being opinionated beyond the concept of enforcing contracts/consent, i.e. people are allowed to structure their economic relationships with one another
23.
by ralusek 2mo ago
I have had production edge cases anticipated by AI before that I hadn't accounted for.
24.
by ralusek 2mo ago
Anecdotal: I used to beg my boss to get some help. I wanted 5 engineers and would have been upset if anyone had been hired that wasn't an engineer until I at least got 1-2. That was up until about 3 months ago. Now, I don't want a
25.
by ralusek 2mo ago
In 65 years Ozempic will have been in use for 65+ years to help people lose weight, but that won’t make it untrue that going for a walk helps you lose weight.
26.
by ralusek 2mo ago
> They want you to rent their compute instead of running it on your own hardware I don't think this is the main thing. I think the real risk open weights AI poses to them is that it allows us to get inference from providers that onl
27.
by ralusek 2mo ago
Alternatively Mitchell and Webb https://www.youtube.com/watch?v=_pDTiFkXgEE
28.
by ralusek 2mo ago
What's interesting is that the way in which they're childish is actually extremely human. In fact, one of the ways that you're often taught to draw more realistically is to stop thinking of the concepts as icons you're d
29.
by ralusek 2mo ago
One benefit of an open source one is that you can, as a large corporation, run it "locally" within your own data center. Even fine tune it.
30.
by ralusek 2mo ago
It's funny that they triggered the infamous "Code Red" moment in OpenAI when the 3-3.1 models came out. I switched to using them for a lot of single-shot LLM calls because they were fast and cheap. Their only area that they w

More ›