Georeferencing Planes Using Claude Code

Governments love PDFs. Not the searchable kind — the scanned, 300-dpi, slightly-tilted kind that laughs at your clipboard.

After years of doing this the hard way, here is my current toolkit:

The toolkit

• Tabula — still the first stop for tabular PDFs that are born-digital.
• pdfplumber — when Tabula gives up, Python usually doesn't.
• Tesseract + pre-processing — for scanned documents; deskewing with imagemagick first saves a lot of grief.
• Camelot — underrated for lattice-style tables.

The truth is that no tool works every time. The fastest workflow is often: try Tabula for two minutes, then move to pdfplumber, then accept that you might be typing.