Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
arresin
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
9 ms
·
91.
▲
by
arresin
11mo ago
- deprecated api use - mixture of styles vs one consistent hand throughout the code. - implicit code vs declarative - convoluted logic when there’s a simpler way. Sometimes complexity just for the sake of it (basically it’s all spaghetti c
92.
▲
by
arresin
11mo ago
This could be 1/4 th the size. LLMs just blab on and on. At least edit the slop throwing it out at the world
93.
▲
by
arresin
11mo ago
Wait the 300 billion has been spent? I thought it was just promised
94.
▲
by
arresin
11mo ago
It’s so weird. Why not vote differently? Is the media captured?
95.
▲
by
arresin
11mo ago
Cleaning up algae doesn’t buy votes
96.
▲
by
arresin
11mo ago
To be fair he’s a proper hard core programmer. I think he’s just gone too much into vibe coding.
97.
▲
by
arresin
11mo ago
The Reddit link is from 18 years ago with people discussing almost the same thing. Damn.
98.
▲
by
arresin
11mo ago
Don’t open time wasting PRs full stop and give oss maintainers a break is the better message to take home from this.
99.
▲
by
arresin
11mo ago
I think he’s resume building.
100.
▲
by
arresin
11mo ago
How is it possible to have this little awareness?
101.
▲
by
arresin
11mo ago
You are under a delusion. I’m serious.
102.
▲
by
arresin
11mo ago
> with his British wife and their 4-month-old baby Ridiculous.
103.
▲
by
arresin
11mo ago
It’s inferior but copilot is even more inferior to it. I used it again recently just to see after cursor And Claude code. It’s laughably bad. Almost like they don’t care.
104.
▲
by
arresin
11mo ago
Please give me the link
105.
▲
by
arresin
11mo ago
Everyone does.
106.
▲
by
arresin
11mo ago
It’s not meant to be anonymous. Obviously.
107.
▲
by
arresin
11mo ago
Is this an ad for brex?
108.
▲
by
arresin
11mo ago
Hey I hope you see this. The scoring needs to be a 0-10 or something with a range rather than pass or fail. Flux one getting the same score for the surfer as Gemini pro 3 reduces the quality of the benchmark.
109.
▲
by
arresin
11mo ago
Are you being sarcastic? If not it’s obvious once you’ve seen it often enough.
110.
▲
by
arresin
11mo ago
That is a very different run in.
111.
▲
by
arresin
11mo ago
> nonsense! Proceeds to talk about baboons.
112.
▲
by
arresin
11mo ago
Thanks for the comment. This was touching to read.
113.
▲
by
arresin
11mo ago
LLMs do take a lot of the joy out of it.
114.
▲
by
arresin
11mo ago
I tried vscode with copilot again after a year of cursor. It’s faster but the LLM aspect is unusable. The diffing is still slow as molasses and the chat is very slow also. LLM wise it’s a joke vs cursor. But it is less laggy (because cursor
115.
▲
by
arresin
11mo ago
Gemini pro 3 was a let down for me too
116.
▲
by
arresin
11mo ago
And then Dario gives an interview on why open source models should be banned due to _____.
117.
▲
by
arresin
11mo ago
Why do so many women have ptsd from dating?
118.
▲
by
arresin
11mo ago
Maybe. For now.
119.
▲
by
arresin
11mo ago
How did it get to this point? Don’t the old folks at Apple use iPhones?
120.
▲
by
arresin
11mo ago
Maybe it’s not a fair judgement because I just looked at the docs, but I didn’t like the syntax on first looking at it.
More ›