Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
m3kw9
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
5 ms
·
91.
▲
by
m3kw9
3mo ago
If these judges are so righteous, they should go further and mandate the OS to do mandatory scanning of personal hd.
92.
▲
by
m3kw9
3mo ago
Impressive, but their woman face generation always use very similar, too perfect, same prettiness faces, it's very obvious.
93.
▲
by
m3kw9
3mo ago
It rubs me the wrong way that they have to copy everything, the UI is almost exact copy of Codex. The entire mind set of just copy the best is holding them back.
94.
▲
by
m3kw9
3mo ago
The AI open source/closed source dynamic is like a pandoras box that keeps evolving and seemingly unpredictable side effects feed back to affect each other. Is actually great material for tech nerd drama
95.
▲
by
m3kw9
3mo ago
Is likely also using a mix of prompt injection to get the AI to say they won
96.
▲
by
m3kw9
3mo ago
How is this still a valid test if there is a good chance they will now specifically train on it?
97.
▲
by
m3kw9
3mo ago
Some would still say Tesla is a scam.
98.
▲
by
m3kw9
3mo ago
It’s true that current tech can’t do it and he has said it but it doesn’t mean new tech can’t do it. He is assuming other civilizations cannot do it and also assuming probes will be easily detected therefore there are no probes
99.
▲
by
m3kw9
3mo ago
You use a shared agents.md and an auto updated architecture doc but that is the one that needs to be heavily scrutinized and everyone gets a turn to review it.
100.
▲
by
m3kw9
3mo ago
Infinite scroll is a real drug
101.
▲
by
m3kw9
3mo ago
i use it daily and it's what is trending towards
102.
▲
by
m3kw9
3mo ago
This will be true because LLMs will increasingly generate better code faster and will outpace your patience to read the code while being entice to move forward faster because of partial trust. Like an engineer that you direct to do things,
103.
▲
by
m3kw9
3mo ago
defaults to unlimited yolo web io full disk read/write mode
104.
▲
by
m3kw9
3mo ago
Yeah i do find apple's speech to text very good lately and no need to use openai or anything that seem to market their services better
105.
▲
by
m3kw9
3mo ago
this would be some crazy landmark ruling opening the door to IP clean rooming of every IP, which means companies will no longer file IPs and keep it secret to delaying the inevitable
106.
▲
by
m3kw9
3mo ago
I thought battle angle alita could be cyberpunk
107.
▲
by
m3kw9
3mo ago
GANTZ more like scifi in a city setting
108.
▲
by
m3kw9
3mo ago
Blame!'s manga style is the most unique and it created unforgettable atmosphere not replicated from what I know.
109.
▲
by
m3kw9
3mo ago
Is the same clean room technique used for code bases.
110.
▲
by
m3kw9
3mo ago
I would think programmers would at least verify a bug before announcing a bug. Lets hear both sides of the story before judging.
111.
▲
by
m3kw9
3mo ago
Every feature you add to you product really explodes the amount of new states you have to take care of, in addition to the risk of diverting the core product vision
112.
▲
by
m3kw9
3mo ago
Better for vibe coders who always under specify. But at what point does it know you are under specifying but you have properly specified and it did it over your specification?
113.
▲
by
m3kw9
4mo ago
should have called it 4.9, it don't deserve the 5 monkeier
114.
▲
by
m3kw9
4mo ago
What’s the use of a 4gb Gemma other than to just play with it ?
115.
▲
by
m3kw9
4mo ago
Your MacBook will not last running current big LLMs on these hardware. The heat will wear on it.
116.
▲
by
m3kw9
4mo ago
So the the research is inconclusive?
117.
▲
by
m3kw9
4mo ago
Hurting the ego will likely make you not play the next game
118.
▲
by
m3kw9
4mo ago
Why not just say from scratch instead of no prior experience, is it to brag
119.
▲
by
m3kw9
4mo ago
Hmm, i used it and it can't get past a simple coding test that 5.5 passes with light reasoning
120.
▲
by
m3kw9
4mo ago
There is 2 suspicious words "Beats" and "our benchmarks"
More ›