-
Vision-model OCR that keeps the layout: what to trust it with
A vision model called a bold italic paragraph 'not bold, not italic'. My OCR now lets the model read, Tesseract place small text, and pixel measurement judge typography: 99% precision on italic words.
-
Translating whole books with local LLMs: the failure modes
My local translation service rebuilds scanned books page by page in another language. Most fixes were not about the model: segment context, glossary scope, a lost-update race and page rebuilding.
-
Six million files, one OSINT shelf: crawl, OCR, embed, NLP
Growing an OSINT records shelf from 35,733 to 185,000+ indexed documents: separate pipelines from the library, files an incremental crawler never sees, 70k image-only PDFs, and OCR measured in page-years.