Launched: PDFreflow - a PDF e-reader whose moat is a layout engine
The problem: reading two-column academic papers or technical PDFs on a phone means constant pinch-zooming. Existing PDF apps are bloated viewers with subscriptions for features nobody asked for; their "reflow" modes (including Adobe's) break on real documents.
The bet: the reflow itself is the product. It's genuinely hard - which is exactly why the incumbents ship broken versions of it and why a solo dev can differentiate there. I spent my time on a custom layout engine (column detection, header/footer stripping, formulas and tables preserved as images) instead of on feature breadth.
The model: free while in beta. After launch: short PDFs (most papers) stay free (permanent funnel for the academic audience that spreads it), long documents will need a single unlock: one purchase covers all documents, any length, forever. Deliberately no subscription: my audience is allergic to them, and "pay once" is itself a differentiator in this category.
Open beta on Android + iOS: https://www.pdfreflow.com/ih
Solo, bootstrapped, Flutter. If it breaks on one of your PDFs, the in-app feedback button sends me the layout - that's what the beta is for.
Have beta users shown that reliable reflow is the reason they would replace their current PDF reader, or is the main signal so far that the engine handles difficult documents well?
Fair question. Praise, yes - several beta readers wrote unprompted that they like it, one specifically about charts and tables reading well in reflow. But nobody has told me they switched readers because of it, and their wish lists (text selection, search, page-by-page reading) say reflow alone won't be enough for everyone. Finding out whether it is - and for whom - is what this public beta is for.
That makes the buyer/use-case question much clearer. If you’re open to it, what’s the best email to reach you on?
a layout engine as the moat is a bold claim in the pdf world - rendering fidelity is where every reader lives or dies. we work the conversion side on swapfile.live (pdf and office formats, all in-browser) and layout is the hardest part of the whole pipeline. what breaks most often in reflow, complex tables or multi-column text?