fardinahsan

8 ms·

1.
by fardinahsan 1y ago
Isn't the sample super biased? StackOverflow is increasingly bleeding users to AI tooling. Shouldn't we expect the remaining users to be increasingly distrustful of AI? I don't use StackOverflow at all anymore.
2.
OpenAI text-embedding-3-large embeddings of the English Dictionary(kaggle.com)
2 points by fardinahsan 1y ago|1 comments
3.
by fardinahsan 1y ago
I love working with LLM text embeddings. The idea of a list of numbers being able to encode semantic information and being able to do math on them is magic to me. Unfortunately, it seems like they are neglected other than being an afterthou
4.
by fardinahsan 2y ago
Could you share the product and some details about it? I'm curious.
5.
by fardinahsan 2y ago
Wut? The causation can flow the other way as well. Having high tike preference results in lower household income. And time preference is probably genetic. They literally controlled for the variable they were testing for....
6.
by fardinahsan 2y ago
Yeah, I'll add a reference to it. Are you watching the repo ?
7.
by fardinahsan 2y ago
I can cook it up if it would actually get some users. Ultimately, I want to make a tool where you can plug in any arbitrary document store/scraper to an LLM with RAG. But I think we are still not there yet in terms of all purpose scrap
8.
by fardinahsan 2y ago
The projects are dead simple. I think if Youtube wanted to, the would have implemented this long back. It's probably a legal landmine and financial tar pit. But thanks nevertheless.
9.
by fardinahsan 2y ago
Yes. I made two recently. - TalkToYoutuber[1]: Download the transcripts from a youtube channel and hook it up to gtp4 with RAG to let me "talk" to youtubers. There's a bunch of youtubers who have useful knowledge to share, bu
10.
by fardinahsan 2y ago
This is overly simplistic thinking. Of Course increases in technology isn't the only factor determining aggregate societal well being or happiness or whatever. But it would be naive and disingenuous to suggest anything other than it be
11.
by fardinahsan 2y ago
Conversely, you can't come up with a proof like this if you don't love math.
12.
by fardinahsan 3y ago
Europe is to Eurasia what the Pacific NorthWest is to North America.
13.
by fardinahsan 3y ago
This is tortured reasoning. Or worse, deliberately misleading phrasing on your part. Let's say that some unspecified statistical technique found two distinct clusters from some given raw data. These two clusters almost exactly line up
14.
by fardinahsan 3y ago
Everytime I read one of these "how to hire" posts, it's always a opinionated clusterfuck of a process that is arbitrary/capricious, gameable and biased towards what OP thinks are good traits. This is a problem that is ve
15.
by fardinahsan 3y ago
The difference is in the magnitude, not kind.
16.
by fardinahsan 3y ago
> Yes, taxes are higher. But so is average quality of life, as opposed to average GDP What does the second part of the sentence have anything to do with creating tech companies?
17.
by fardinahsan 3y ago
Are you making a quantile-quantile comparison? It's probably likely that you are making way more than the average romanian dev and the same is likely to apply in norway.
18.
by fardinahsan 3y ago
Can you go in some high level details as to why it was slow and what you did to make it fast. That's always the most interesting part of a post like this.
19.
by fardinahsan 3y ago
The reason is quite straightforward. The survival of the human species depends on it being this way.
20.
by fardinahsan 3y ago
I don't think saying you have a bone to pick with "AI" is doing your message any favors. What are you specifically against LinearRegression? XGBoost? ResNet? LLMs? AlphaFold?
21.
by fardinahsan 3y ago
Because "creating memories" can't be exchanged for social status.
22.
by fardinahsan 3y ago
Let's assume it were a problem big enough to address, then what?
23.
Ask HN: How would you solve good candidates getting rejected by ATS
1 points by fardinahsan 3y ago|3 comments
24.
by fardinahsan 3y ago
The scientists had it coming for using excel.
25.
by fardinahsan 3y ago
And university essays are effective for that?
26.
by fardinahsan 3y ago
"Teens" seems to be an euphemism.
27.
by fardinahsan 3y ago
I hope the incoming overabundance of GPT generated papers at least as a side effect gives the scientific community a wakeup call. If publishing becomes a non prestigious enough thing because anyone can GPT can do it, then at least those who