1
1 Comment

The pdf thing nobody tells you is annoying until it happens to you

so i was helping a friend put together a deck. she had this old pdf report, like 40 pages, and buried in there were maybe 15 charts and photos she needed to reuse. nothing fancy, she just needed the images out of the pdf and into a folder.

i said sure, i will do it real quick.

i was wrong about the "quick" part.

first thing i tried was just... screenshotting each one. worked, but every image came out slightly blurry and cropped wrong because i kept catching a sliver of the page margin. after the fourth screenshot my hand was already tired of the same crop-drag-save loop.

so i looked for an actual tool. first result wanted an email and a "create account" before it would even open the file. second one opened the pdf fine, extracted the images fine, then handed them back at like a third of their original resolution — completely useless for a deck. third one flagged the file as "too many pages, upgrade to continue" even though it was a normal 40-page report, not some giant scanned archive.

at that point i just sat there thinking, this has to be one of the most common things people do with a pdf, and somehow every tool treats it like a premium feature.

that is the annoying part about this kind of problem — it's not hard, technically. pulling embedded images out of a pdf isn't some deep computer science puzzle. it's just tedious enough that nobody wants to build it well, so it sits there half-solved by ten different sites that all want something from you before they'll help.

so yeah, i ended up building it. drop your pdf in, it finds every image inside — full resolution, no recompression, no watermark stamped across the corner and gives them back to you as a zip. no account, no page limit games, no "upgrade to actually get your own file back."

that is basically the whole idea behind the pdf image extractor on kubergaent. you give it a pdf, it gives you the images. that's it. that's the feature.

if there is a pattern i keep noticing with these little projects, it's this: the problems worth solving aren't the exciting ones, they're the ones everyone quietly deals with and nobody bothers to fix properly. extracting images from a pdf. converting a clip to a gif. small, boring, extremely common. that's usually a better signal than some flashy idea nobody's actually run into yet.

if you've got a pdf full of images you need out, and you don't want to sign up for anything or watch your resolution get destroyed,kubergaent has it.

what's your version of this? some small annoying task you keep doing by hand because every tool for it is somehow worse than just doing it yourself?

on August 13, 2026
  1. 1

    You're noticing the pattern yourself ("small, boring, extremely common"), so one push: you're not building tools anymore, you're building a repeatable machine. Video-to-gif, now pdf-image-extract, same playbook, find a micro-task everyone does and every tool ruins with signups and paywalls, ship the clean version. Not a project, a factory, worth more than any single tool in it.

    Which reframes it: ten scattered free tools is a hobby. Ten under one roof, one brand, one "never make you sign up, never wreck your file, never paywall the obvious" promise, is a destination people bookmark. The tools are commodities; the trust across them compounds. Unify them.

    What's the through-line that makes someone who used the gif tool come back for the pdf one?