Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
adityamwagh
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
4 ms
·
31.
▲
by
adityamwagh
5mo ago
It’s a huge memory hog. The other day it was using 2.5GB!
32.
▲
by
adityamwagh
5mo ago
What’s with the all the hate for AI assisted writing on HackerNews? It’s a tool and people use tools all the time. It saves TIME and helps in improving coherence of one’s articles.
33.
▲
by
adityamwagh
5mo ago
Taalas is a nice concept, but I don’t want to use the same model forever!
34.
▲
by
adityamwagh
6mo ago
> We also found evidence that models that have seen the problems during training are more likely to succeed, because they have additional information needed to pass the underspecified tests. No shit, Sherlock!
35.
▲
by
adityamwagh
6mo ago
GitHub was down too! Its uptime has been so bad recently.
36.
▲
by
adityamwagh
6mo ago
And somehow he’s worth $655B. Pays to lie I guess haha.
37.
▲
by
adityamwagh
6mo ago
It eventually will. But OP never asked about M5 specifically.
38.
▲
by
adityamwagh
6mo ago
Yes, see https://asahilinux.org/
39.
▲
by
adityamwagh
6mo ago
There's not enough internet-scale data for robotics. The gap is huge! So anyone that claims to have a GPT like model is not behing honest.
40.
▲
by
adityamwagh
6mo ago
Yes there should be. But there won’t be until US stops lobbying and American public elects lawmakers that work for people instead of their own pockets.
41.
▲
by
adityamwagh
6mo ago
I think enterprise users have a API for the whole company do pay as you go? I believe that’s where they make the most money.
42.
▲
by
adityamwagh
6mo ago
Isn’t that actually a valid way to test? IMHO Performing under pressure is a capability signal in itself.
43.
▲
by
adityamwagh
6mo ago
> It's usually more dangerous for women to give birth lying down, so why do they? It's all because of a Frenchman who decided it was more convenient – for men. The first paragraph itself makes me not want to read further.
44.
▲
by
adityamwagh
6mo ago
Do we even need documentation with LLMs? :)
45.
▲
by
adityamwagh
6mo ago
How would they make money from the tokens then haha? The main revenue driver of these companies is to get people to use more tokens. That’s what they will optimise for. Getting the developers out of the way is the way to do it.
46.
▲
by
adityamwagh
6mo ago
You need to get in touch with them personally. I feel all hiring happens in circles.
47.
▲
by
adityamwagh
7mo ago
My favourite one is Iosevka Extended.
48.
▲
by
adityamwagh
7mo ago
Yeah, I kinda expected it. I'm a gruvbox and catpuccin guy haha.
49.
▲
New VSCode Default Themes
(code.visualstudio.com)
2 points
by
adityamwagh
7mo ago
|
3 comments
50.
▲
by
adityamwagh
7mo ago
In the latest version of VSCode, the default themes were updated.
51.
▲
by
adityamwagh
7mo ago
That’s half a day worth of revenue for Meta. Why don’t companies get fined billions?
52.
▲
by
adityamwagh
7mo ago
In it's current form, homebrew is amazing. It's not that slow and recent updates have made it really good to use. May I know the reasons for a Rust rewrite?
53.
▲
by
adityamwagh
7mo ago
Yup! I'm thinking of ways in which I can use their iroh library to help with some of the workloads at work ( https://neuracore.com , https://github.com/NeuracoreAI/neuracore )
54.
▲
by
adityamwagh
7mo ago
You're welcome. Great thing about it is it's a just a binary that you can fetch using curl anywhere, and then do ./sendme send <file/folder> and on the other machine type the command sendme tells you to in the term
55.
▲
by
adityamwagh
7mo ago
Love the folks from n0. I regularly use their sendme cli for peer to peer file transfer!
56.
▲
by
adityamwagh
8mo ago
Check out raindrop.io, I use it everyday it’s it’s been a pleasure to use.
57.
▲
by
adityamwagh
9mo ago
This account was created one hour we should ignore this comment. :)
58.
▲
by
adityamwagh
9mo ago
This just post just seems like a conspiracy theory.
59.
▲
by
adityamwagh
9mo ago
https://adityamwagh.me/
60.
▲
by
adityamwagh
1y ago
It’s a very popular article that has been around for a long time!
More ›