Notes on reading research papers,
and building the app for it.
Plot digitization, local-first AI, the comparison shopping behind building yet another PDF reader — and what we learned shipping a Mac app to researchers.
-
How to extract every figure from a research paper
A practical look at the figure-extraction problem — why naive approaches fail, what works in practice, and how to build a usable 'all figures at a glance' view.
Read → -
The 6 PDF readers I tried before building my own
Notes from a year of trying every tool aimed at researchers — what each got right, what each got wrong, and what was missing across all of them.
Read → -
Why local-first beats cloud for research papers
Sending your PDF to a server to ask 'what does this paper say?' sounds convenient until you think about what that server now knows about you. A case for on-device reading.
Read → -
How to digitize a plot from a PDF (without losing precision)
A practical guide to recovering the underlying numbers from a figure in a research paper — calibration, picking, and the precision tradeoffs that nobody warns you about.
Read →