The AI hardware race is still going strong, but with so many rapid changes to the fundamental architectures, it doesn't make sense to bet everything on specialized hardware just yet.. It's happening, but it's expensive and slow.
There's just not enough capacity to build memory fast enough right now. Everyone needs the biggest and fastest modules they can get, since it directly impacts the performance of the models.
That's when you go a level deeper and have every template use another template (e.g. a Helm subchart or Helm library) only to realize scoping and templating is completely fucked in Helm.
I've wanted to make something like this myself, so thanks and good job!
How does this work? Does it rely on GPT to extract the data or does it actually generate a bunch of selectors? If it's the former, then the results aren't reliable since it can just hallucinate whole results or even just parts.
I haven't put together a good test framework yet, but qualitatively, the results are surprisingly good, and hallucinations are fairly low. The prompt tells GPT to say (not available) if needed, which helps.
I'm going to try the "generate selectors" approach as well. If you'd like to learn more or discuss just reach out via email (marcell.ortutay@gmail.com) or discord (https://discord.gg/mM54bwdu59 @ortutay)
Just put them on a train during work hours! We have really good coverage here but there's congestion and frequent random dropouts, and a lot of apps just don't plan for that at all.
"Permissions" seem too specific a term to use as a general term. It's something I'd use to describe the specific rights a role may have in role-based access control, and not authorization as a whole. I'll stick to authn/authz for abbreviations, auth for both or if it's not specific, and if it's for documentation or cross-department communication I'll just write the whole word.
It also can't fetch more than 2 pages of results for either web or image search. Reverse image search just straight up doesn't work. I don't know why they have it.
With searxng, searching at least works. It also has plugins that I think can do some of what Kagi is useful for (but it depends on the instance)
> It also can't fetch more than 2 pages of results for either web or image search
I can't speak for image searching, but I found when I get limited results from Kagi, The other search engines that produce More results just pad those results with noise. Sometimes I'll do a specific search and have no results on Kagi, so I switched to another engine justified I have thousands of unrelated results
> The other search engines that produce More results just pad those results with noise.
Yeah, that's been an increasing problem.. Google has been especially bad with this for a while now.
When I tried to search for something where I wasn't sure exactly what I was looking for, I could only change my query so many times until it stopped being what I wanted.
I came here to suggest Kagi. I bloody love it. No ads, ability to remove or downgrade or promote particular sites, lenses, I love it all.
Your comment - honestly, hand on heart, I can’t remember the last time I paged through beyond the first page of any search engine. I didn’t even know Kagi had this “problem” and so I guess I just don’t see it as a problem. But then I realised I’m absolutely in that bit of “must be on first page” internet lore, too…
> I didn’t even know Kagi had this “problem” and so I guess I just don’t see it as a problem.
Took me a while before I actually realized this limitation as well. I started noticing it when I was searching for things I didn't know exactly how to find.
Kagi is great for when you know exactly what to find and have a general idea of where, but when you actually need those n-page results to help you refine your search, and you can't, it's a showstopper for me.
I'm not even sure why it's a limitation. It might not be intentional?
Good point! Maybe that explains why they chose it only for the bands. People touch the phone the most at the outsides, so this sounds like a great trade-off between allowing the phone to cool itself (via the aluminum back) while not being hot or cold to the touch when holding it.
Well, the predecessor is stainless steel, which I imagine is closer to the Ti number, and in my experience with 12/13 Pro it's always the back that throws off heat, not the frame.
Anyhow, just the frame is this material, inside its Al.
There's just not enough capacity to build memory fast enough right now. Everyone needs the biggest and fastest modules they can get, since it directly impacts the performance of the models.
There's still a lot of happening to improve memory, like the latest Titans paper: https://arxiv.org/abs/2501.00663
So I think until a breakthrough happens or the fabs catch up, it'll be this painful race to build more datacenters.
reply