Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
serbuvlad
searching Neon…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
6 ms
·
1.
▲
by
serbuvlad
3d ago
I find that the people who are most concerned about circular financing in the case of AI, are those who fearmonger about deflation and talk about the importance of monetary velocity for the economy. Isn't this a contradiction?
2.
▲
by
serbuvlad
5d ago
1. I don't know, Transformers seem to be pretty good at those Domain specific skills. 2. I'm sure mental gyms will become common, although I'm not sure if chess is a very good mental gym. It may take a hundred years or so, th
3.
▲
by
serbuvlad
5d ago
This is only the case for the Industrial Revolution. There have been many other thinking technologies and they ALL have been attacked in exactly the same way. * Socrates attacked the invention of writing, arguing it would make people intell
4.
▲
by
serbuvlad
5d ago
I'm not saying this situation is different from all of the other technological paradigm shifts in history. But what concerns me is that authors of articles such as this feel no need to differentiate themselves from other anti-technolog
5.
▲
by
serbuvlad
6d ago
It's just a local minimum. Things today are generally pretty good, and work generally well. The only issue is that things work well all in the same way, and deficiencies are also shared. Things will improve as creatives try far-out ide
6.
▲
by
serbuvlad
8d ago
Apple and Mistral are not in the same market. Siri is a minor part of the Apple ecosystem. Fundamentally, it can call into a better LLM provided by a better company. Siri's only utility is that it has access to your iData and can contr
7.
▲
by
serbuvlad
9d ago
Sure, but SIMD extensions and base ISA solve different problems. As for GPUs, while the ISA is not constant, it's STILL C-ish running over a dynamic hardware scheduling layer. Edit: To clarify, I have nothing against a closed ISA, I ju
8.
▲
by
serbuvlad
9d ago
C is a low-level language for the current ISAs we have, though not for Itanium. So the question is if we really want lower level ISAs. Probably not? There are many ways in which our current ISAs are actually thoughtfully optimized for super
9.
▲
by
serbuvlad
9d ago
I've been thinking about this as well. Certainly we will have mental gyms in the future.
10.
▲
by
serbuvlad
10d ago
Ethics aside, you can create an NGO with three people and do a lot of stuff in it's name that you can't do as a private individual or it would be a lot more expensive to do. If everyone did this, it would stop being a special abil
11.
▲
by
serbuvlad
10d ago
Places that have lots of trash around will vandalize an unsupervised robot. Places that will leave alone an unsupervised robot don't have lots of trash around. Entirely disjoint sets.
12.
▲
by
serbuvlad
11d ago
114M a year buys you only 650 people? Damn. I really should be moving to America at some point :)
13.
▲
by
serbuvlad
11d ago
People need to learn to use .git/info/exclude WAYYY more! You use Cursor which creates .cursor? Cool! Put it in .git/info/exclude. No need to pollute the project .gitignore with that. .gitignore is for artifacts that ari
14.
▲
by
serbuvlad
11d ago
Cloud gaming is incredible tech, I used GFN for a while and it was incredible. HOWEVER, I will inevitably want to play weird stuff like "milk inside a bag of milk inside a bag of milk", Pseudoregalia or Civ V Lekmod, which these c
15.
▲
by
serbuvlad
11d ago
What's the performance delta? I'm waiting on a SM reservation but I wouldn't mind saving $500+ :) Is there any "jank" during active use or is this mostly a set up thing?
16.
▲
by
serbuvlad
13d ago
To me all this makes the label of AGI completely meaningless. What AGI has always meant (eg. in 2019) is Artifical General Intelligence. Artificial -- something made by humans instead of occurring naturally General -- not confined by specia
17.
▲
by
serbuvlad
17d ago
Obviously some variation of this will happen again, it will kill someone* and then LLMs will become massively regulated. Just like every technology ever in our history. It does seem like AI is perfectly controllable given how much it is use
18.
▲
by
serbuvlad
20d ago
You can certainly nudge it directionally, but it has its own style. And verbal tics etc. So it simply cannot keep documentation or a codebase in the same style, and it cannot write <information> in <style>. With 3.5 Turbo I us
19.
▲
by
serbuvlad
20d ago
I remember the most interesting things to do with 3.5 Turbo was to ask it to rewrite paragraphs in different styles. Why can modern LLMs basically not do that anymore? They each have their own very particular writing style? It it all just b
20.
▲
by
serbuvlad
21d ago
I have resigned to the fact that there be dragons. I can tell you how things are with Sol. With Sol you need to know what you're doing and roughly how you'd implement it by hand yourself in order to use Sol effectively. With Sol y
21.
▲
by
serbuvlad
21d ago
To be fair LLMs are very good at information synthesis jobs, such as finding bugs.
22.
▲
by
serbuvlad
21d ago
* it's artsy, which fits Apple's PR image * it's critically acclaimed (86 metacritic, 10/10 IGN) * whatever person decided this likely knows nothing about video games * most importantly it's a modern game in UE5 tha
23.
▲
by
serbuvlad
22d ago
No why would I do that? :)
24.
▲
by
serbuvlad
22d ago
I find this very interesting in 2026. I use my gmail address for everything and I don't really get spam (except from services I've subscribed to legitimately but haven't bothered to configure to not send promotional mail). I
25.
▲
by
serbuvlad
23d ago
Yes but as a dev you are shaped by the projects you work on.
26.
▲
by
serbuvlad
23d ago
I am talking from the perspective of the individual developer. You don't need your impressive product to succeed to land a good career. If Alice is doing LLM-from-scratch work today, and Bob is doing agent harness work today, Bob'
27.
▲
by
serbuvlad
23d ago
Sure, but I'm sure most of their lead devs are well paid now. In general C++ work and similar, if not in Chrome development.
28.
▲
by
serbuvlad
23d ago
If you learnt to build a browser in 2000 you're probably doing well for yourself lol. Not a lot of demand, but also probably not a lot of supply.
29.
▲
by
serbuvlad
24d ago
Asking for free inference is like asking for free gold in today's economy. :)
30.
▲
by
serbuvlad
25d ago
Personhood is poorly defined.
More ›