Post

aly
@aly.codes

so you’re saying you need software that is natively multimodal, good at conversions, can work in iterative verification loops, and is good at navigating complicated proprietary formats with intuitive prodding? Hmm

seed paste soy goblin@phasebinary.bsky.social

"we don't need fire, we just need a way to cook our food"

aly
@aly.codes

I highly suggest reading my post back to yourself out loud

0 reposts4 likes

Kimmy W
@ghostynewt.bsky.social

pdf conversion should be a oneshot command like docling or allenAI’s olmOCR, I don’t want to go back and forth with an agent on a task like that i agree with your other points though, sorry for being annoying & not getting it

1G-N15 ⬡
@1gn15.com

I think what aly's trying to say is that such a command line tool most likely uses AI in the backend to convert it back into docx. AI usage doesn't have to be chatbot-style.

norvid_studies
@norvid-studies.bsky.social

further agreeing with everyone here, while "do everything perfectly on the first try with no specifications" is a laudable goal, in the unlikely event that falls short, "debug" it in conversational english is the best format for further feedback for a random user rather than a scripting language