xgulfie

7 ms·

91.
by xgulfie 4mo ago
Many many HN posts are not news
92.
by xgulfie 4mo ago
what makes you think AI won't be the ultimate rent-seeker
93.
by xgulfie 4mo ago
Relying on users to use an LLM to generate their own docs presupposes that the users have a Claude subscription or whatever. That sucks imo
94.
by xgulfie 4mo ago
So if they say "idk Claude did it", what would you write in the PR review box?
95.
by xgulfie 5mo ago
Art is definitely being devalued by AI, today. I'm sure Banksy is fine but artists along the margins are competing with genAI
96.
by xgulfie 5mo ago
To your other point, skill retention is important for me, the person typing this, because we live in a society where if I don't develop my skills I become unemployed and die in a ditch
97.
by xgulfie 5mo ago
I anticipate that heavily using AI to do the dev work for you for a few years could very well be worse than just taking the years off, since you're training your brain that you can just reach for the AI
98.
by xgulfie 5mo ago
> AI has nothing to do with laziness or greediness.
99.
by xgulfie 5mo ago
If you can't see greed in the LLM sphere you are not looking very hard.
100.
by xgulfie 5mo ago
LLM content is so exasperating to read, it always reads like a student trying to pad out their paper, or like a press release with no details
101.
by xgulfie 5mo ago
large language models do not have pigeon-level intelligence. They can't even feed themselves.
102.
by xgulfie 5mo ago
What time is AI going to free up for me? Can AI go to the grocery store for me, do my laundry, do my dishes? Can it let me clock out early? The spoils of AI do not go to individuals
103.
by xgulfie 5mo ago
And they all want to auto-update, too.
104.
by xgulfie 5mo ago
saving you a click: it's Radioactivity
105.
by xgulfie 5mo ago
This is being downvoted but it's worth noting if only for the "be careful" aspect. That said, we need more people distilling models IMO, just be ready for a C&D and a ban
106.
by xgulfie 6mo ago
If I'm writing JavaScript, the cognitive overhead of doing async is less than dealing with nodejs require() and import syntaxes and module compatibility
107.
by xgulfie 6mo ago
Oh thank God I can still give Elon Musk more money and power but don't have to read about politics
108.
by xgulfie 7mo ago
What is it that Zuck called people who trusted him? Oh right
109.
by xgulfie 7mo ago
When I first look at this I think "hey it would be nicer if it wasn't falling apart", but you could argue that's kind of the point. Well done
110.
by xgulfie 7mo ago
The author mentions urban decay and dilapidation multiple times and very clearly worked that into the design here
111.
by xgulfie 7mo ago
You're absolutely correct! Having the LLM using more tokens does improve its output. Here's why this works: ## More tokens = smarter outputs When an LLM uses tokens, it is putting more information into its context ## Better cont
112.
by xgulfie 7mo ago
It's because we only really know one economic system but we've known many people
113.
by xgulfie 7mo ago
It must be intentionally obtuse, nobody could ever confuse copilot for copilot
114.
by xgulfie 7mo ago
OBS + kdenlive can also get you here, but this product is meant to be purpose-built and time-saving specifically for the computer instructions use case
115.
by xgulfie 7mo ago
Funny how people are so critical of this and yet fawn over TOON
116.
by xgulfie 7mo ago
Ah so obviously making the LLM repeat itself three times for every response it will get smarter
117.
by xgulfie 7mo ago
[flagged]
118.
by xgulfie 7mo ago
LLM quirks are not something all humans have been experiencing for thousands of years
119.
by xgulfie 7mo ago
Thinking about music mathematically is at least a good way to understand music
120.
by xgulfie 7mo ago
But build steps suck

More ›