Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
dakolli
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
13 ms
·
211.
▲
by
dakolli
4mo ago
We have what, 2500 of them, and we threaten them with them... Wouldn't you? Iran deserves them too.
212.
▲
by
dakolli
4mo ago
That sounds like a Thiel funded project, given the Tolkien name.. If so, I won't ever be using Pi. edit: From first glance, it doesn't look like it. But I basically don't trust any tech company that takes to Tolkien naming co
213.
▲
by
dakolli
4mo ago
This is a literal Evil (I mean literally) idea wrapped up with statistical terminology and made to seem "logical". You should re-evaluate this.
214.
▲
by
dakolli
4mo ago
What's your source? The Bureau of labor statistics that is marking most of unemployed people as "not seeking employment" because they aren't walking into workforce centers?
215.
▲
by
dakolli
4mo ago
You don't have to give them a choice in the model, you can set your app to use kimi 2.6 only.
216.
▲
by
dakolli
4mo ago
I don't really have an issue with kindle, I just download anything off of z-library and use the email to kindle tool. I haven't given Amazon a penny other than from the original purchase of the kindle. I only have used epub/p
217.
▲
by
dakolli
4mo ago
Why don't you do sign in with openrouter oauth, then users can create an account and assign a key to your app, with spending limits. It's trivial, plus if you get a decent amount of users your app will appear on the leader board a
218.
▲
by
dakolli
4mo ago
If you have 80k in hardware you can run it.. There is not such thing as an effective local model that runs on consumer hardware, anybody telling you otherwise is lying, delusional. JuSt a FeW MoRe ReLeAsEs
219.
▲
by
dakolli
4mo ago
I'll do better fo sho
220.
▲
by
dakolli
4mo ago
Lmao, imagine doing all this instead of just centering the div yourself. This explains why everyone I work with that uses llms is slow af. Its all pseudo-productivity, its kinda similar to how people sit around customizing their nvim config
221.
▲
by
dakolli
4mo ago
Loop engineering, whatever that is, is obviously just a way to get people to increase the amount of tokens required per task/request. They did the same thing with Ralph loops, they just need more revenue. Just write your code and use i
222.
▲
by
dakolli
4mo ago
I see three possibilities. 1. Likely: This is a completely contrived marketing stunt. Release a spooky sounding model, market it as such, then use the narrative to regulate open source models and dig your regulatory moat. Notice the emphasi
223.
▲
by
dakolli
4mo ago
Why have you sent this same message multiple times?
224.
▲
by
dakolli
4mo ago
In America it's more important to maintain or grow shareholder returns than it is to not poison your customers. Its perfectly alright to give your customers cancer (Roundup, Fracking, Massive amounts of sugars which lead to metabolic d
225.
▲
by
dakolli
4mo ago
How else are they going to justify giving out this gigantically profitless model? They must train on your data on the premise of safety.
226.
▲
by
dakolli
4mo ago
This is a problem with people with LLM psychosis who now think they have superpowers, they are completely unaware and just do things naively, they've lost all ability to think for themselves. The LLMs that are thinking for them certain
227.
▲
by
dakolli
4mo ago
No one is training their models on their closed sourced proprietary code. They own github, why would they need to do this.
228.
▲
by
dakolli
4mo ago
There's literally no indication that this is the case, or will ever be. Unless you're a completely naive person who's impressed with all output of an LLM because you don't know what you're talking about. These model
229.
▲
by
dakolli
4mo ago
I suspect it just sucks, these models aren't useful. Stop lying to yourself.
230.
▲
by
dakolli
4mo ago
This is all marketing, you don't have to believe everything a company is saying about themselves, and you shouldn't. Although, I could see Anthropic making a model purposely dangerous so there are bad outcomes and they can use tha
231.
▲
by
dakolli
4mo ago
I'm happy not using llms because I like learning things and working hard. I love writing code, it's genuinely my favorite thing thing to do. Using llms is the equivalent of driving to the store that's 3 blocks away, just like
232.
▲
by
dakolli
4mo ago
Wasn't it leaked in the Claude Code source that it was all regex?
233.
▲
by
dakolli
4mo ago
They're useless tools only helpful to lazy people that don't want to learn by themselves.
234.
▲
by
dakolli
4mo ago
best comment I've seen on this site in a while
235.
▲
by
dakolli
4mo ago
Its an impressive project. You could get really far by just picking a softer color palette. There's an over reliance on color coding going on that's hard on the eyes. Optional pagination would be nice, I would use a Zebra pattern
236.
▲
by
dakolli
4mo ago
LLM UI/UX is so bad. Like how hard is it to put a weekends effort into UX for something like this.
237.
▲
by
dakolli
4mo ago
I agree, but many do not. Anyways, those 22k people are going to want to be on both regardless.
238.
▲
by
dakolli
4mo ago
You don't need an llm or a human to handle 99% of account recovery tickets. They're just shoving LLMs into everything because decision makers have llm psychosis because openclaw modified their calendar a few times and now think th
239.
▲
by
dakolli
4mo ago
What are their alternatives? I'm assuming that most of the 22k people with large audiences that need a large platform to reach them. Meta unfortunately, is the only platform that allows people to reach across many demographics. Its the
240.
▲
by
dakolli
4mo ago
I will not be spending thousands in hardware to run the worlds most mediocre llms at meh speeds. Sorry. I know for llm bros they think every output made by an LLM is magic, like every NFT guy thought every NFT collection was game changing,
More ›