That was my thought. Young kids fresh out of school are really easy to manipulate into delusions of grandeur, especially when said delusions are offered by the richest person in the world. He's gonna leave them out for the wolves.

save

path: 0 14947794 14948083, hotness: undefined, score: 56, children: 3

SnotFlickerman 47 points a year ago

Either that or Musk himself is truly so incompetent he thinks these kids are true geniuses. Honestly, with how things are going, that's a fiddy-fiddy chance, because Musk is somehow almost as unbelievably stupid as Trump.

save

path: 0 14947794 14948083 14948126, hotness: undefined, score: 47, children: 2

BartyDeCanter 18 points a year ago

Why not both?

save

path: 0 14947794 14948083 14948126 14948734, hotness: undefined, score: 18, children: 0

lka1988 9 points a year ago

You're probably not wrong, what with him awkwardly hopping around onstage at multiple trump rallies.

save

path: 0 14947794 14948083 14948126 14950281, hotness: undefined, score: 9, children: 0

Opisek 54 points a year ago

Technically OCR is an application of machine learning.

Not an LLM, though.

save

path: 0 14950637, hotness: undefined, score: 54, children: 3

SwordInStone 15 points a year ago

A world of difference

save

path: 0 14950637 14952065, hotness: undefined, score: 15, children: 0

jsomae 2 points a year ago

regularizing the OCR'd form into a json/html file might be a good application of an LLM though. Perhaps this is what they were asking about.

save

path: 0 14950637 14952314, hotness: undefined, score: 2, children: 1

kureta 2 points a year ago

I doubt they even know what they are asking about?

save

path: 0 14950637 14952314 14969003, hotness: undefined, score: 2, children: 0

tsiad_mordecai_miktros 49 points a year ago

yes me send me what you want me to parse and i will get back to you in 3-4 business days

save

path: 0 14947022, hotness: undefined, score: 49, children: 7

eldavi 20 points a year ago

be sure to include the metadata too. lol

save

path: 0 14947022 14947093, hotness: undefined, score: 20, children: 6

OwlPaste 15 points a year ago

And processing fee!

save

path: 0 14947022 14947093 14947101, hotness: undefined, score: 15, children: 5

tsiad_mordecai_miktros 4 points a year ago

the only fee i want is pics

save

path: 0 14947022 14947093 14947101 14947151, hotness: undefined, score: 4, children: 4

DeathsEmbrace 4 points a year ago

save

path: 0 14947022 14947093 14947101 14947151 14950214, hotness: undefined, score: 4, children: 3

mesamunefire 47 points a year ago

Imagine getting a job like this and now half the nation knows your name...thats terrifying. being an intern may mean you have no idea of the true scope of what they are asking you to do.

save

path: 0 14947739, hotness: undefined, score: 47, children: 5

Pieisawesome 22 points a year ago

They are public employees who are changing things at the core of our government. Why wouldn’t we know their names?

Government employees names aren’t secret (asides from a few exceptions) nor is their pay

save

path: 0 14947739 14951520, hotness: undefined, score: 22, children: 0

GrumpyDuckling 18 points a year ago

We know that his dad is an engineering professor at university of Nebraska too. Really calls into question his credentials. I checked the other day and they had already removed his contact info from their website.

save

path: 0 14947739 14950227, hotness: undefined, score: 18, children: 0

crusa187 15 points a year ago

Yeah, seems that’s the point. Old enough to competently perform what they’re told, but too young to realize the gravity of the situation and how wrong it is to partake in it.

save

path: 0 14947739 14952067, hotness: undefined, score: 15, children: 2

eldavi 16 points a year ago

that's why we have 18 year soldiers ....

save

path: 0 14947739 14952067 14952083, hotness: undefined, score: 16, children: 1

ChapulinColorado 10 points a year ago

It’s ok, with the experienced gained from being forced to grow up, some will come home and use their savings to buy a dodge ram on a 7 year loan at 18% apr.

save

path: 0 14947739 14952067 14952083 14952699, hotness: undefined, score: 10, children: 0

NicolaHaskell 47 points a year ago

This is that special blend of Tablet Kid "I don't need to know things I can google them" and Rich Kid "I don't need to do things I can crowdsource them" that makes for that Distinctively VP "I don't know what I'm doing and nobody can tell 👈😎👉"

save

path: 0 14952355, hotness: undefined, score: 47, children: 0

CapriciousDay 25 points a year ago

save

path: 0 14947829, hotness: undefined, score: 25, children: 0

tetris11 22 points a year ago

I have to admit, PDF parsing being such a hot and profitable topic in computer science was really something I never saw coming.

PDFs? The things you can select text from? And when not, there's decent OCR? And when not, you just ask the person to send you an email or a word doc?

It sounds like LLMs are looking for a new unpolluted source of historical data that they can learn from, and this source exists in the form of old scanned-in paper documents. That's the only reason I can fathom as to why this is such a big thing now.

save

path: 0 14955638, hotness: undefined, score: 22, children: 4

chicken 9 points a year ago

Every time I try to convert a PDF to epub or something, or OCR one that doesn't actually have selectable text, it turns out shit. I assume the real reason people would want to get LLMs involved is that there is actually a lot of ambiguity in what a correct conversion would be, and there are a lot of PDFs out there.

save

path: 0 14955638 14957157, hotness: undefined, score: 9, children: 1

JustAnotherKay 2 points a year ago

I self host sterling-pdf and I haven't had an issue with file conversion in... When did I set this thing up?

To be truthful, the machine I had it running on has been sent to the grave (I sold it) so I don't actually have this service running right now

save

path: 0 14955638 14957157 14967615, hotness: undefined, score: 2, children: 0

MonkderVierte 5 points a year ago

Selecting text doesn't work in most multi-column pdfs and good OCR cost money. And if the original source is lost and you want an exact copy in word, the OCR tools need to be really good at guessing whitespace-to-line ratio, because pdf is only an output format and not a processing format.

For most other converting needs, there's pandoc, imagemagick and ffmpeg.

save

path: 0 14955638 14966460, hotness: undefined, score: 5, children: 0

sudo 5 points a year ago

Training the most insane AI model on classified federal documents.

save

path: 0 14955638 14956169, hotness: undefined, score: 5, children: 0

GissaMittJobb 19 points a year ago

Is this fake?

For context, this is the guy who figured out how to see what's written on some ancient Greek Scrolls without destroying them. It seems slightly far-fetched that he wouldn't know better.

save

path: 0 14948237, hotness: undefined, score: 19, children: 6

SwordInStone 6 points a year ago

What Greek scrolls?

save

path: 0 14948237 14948370, hotness: undefined, score: 6, children: 5

GissaMittJobb 11 points a year ago

Ok so they were apparently in Greek but not from Greece. Source: https://news.unl.edu/article-2

save

path: 0 14948237 14948370 14948685, hotness: undefined, score: 11, children: 4

Telorand 10 points a year ago

Even granting the quality or contribution he had to that effort, there's a huge difference between, "I can make a computer read burnt scrolls," and, "I can make government software with industry-standard protocols and security."

By way of comparison, just because I can write automation software for my company's apps doesn't mean I could just jump into doing Linus Torvalds' job maintaining the Linux kernel.

save

path: 0 14948237 14948370 14948685 14950838, hotness: undefined, score: 10, children: 3

GissaMittJobb 11 points a year ago

Sure, but the difference between "I win awards for figruing out how to decipher ancient scrolls that no one has been able to do despite their best efforts" and "I can't research well enough to discover appropriate existing tools for document conversion" is very hard to reconcile.

save

path: 0 14948237 14948370 14948685 14950838 14951001, hotness: undefined, score: 11, children: 2

rustydrd 17 points a year ago

Ah yes, the famed document <-> JSON converter.

save

path: 0 14954911, hotness: undefined, score: 17, children: 1

xthexder 12 points a year ago

It's so easy! Watch:

{"contents": "<garbled .docx contents goes here>"}

save

path: 0 14954911 14955485, hotness: undefined, score: 12, children: 0

cmgvd3lw 15 points a year ago

Actually this is what they do.

save

path: 0 14947341, hotness: undefined, score: 15, children: 2

some_guy -5 points a year ago

Yeah, I don’t really get this one. The class clown is the kid who recognizes a function of a tool, correctly at that. Unlike a dipshit lawyer who let it hallucinate bogus case law. Hilarious.

save

path: 0 14947341 14948070, hotness: undefined, score: -5, children: 1

masterofn001 28 points a year ago

There are programs that exist that explicitly do these sorts of things.

They have been around since long before llm.

This is a lazy and uneducated question.

He demonstrates he has done zero research and goes straight to the buzzword because he knows nothing.

save

path: 0 14947341 14948070 14949307, hotness: undefined, score: 28, children: 0

ininewcrow 11 points a year ago

Just use Deepseek for US government data .... what could go wrong?

save