Show HN: PDF to Markdown that preserves layout, images, and tables

I built this because I often need clean Markdown from PDFs for writing, documentation, and LLM workflows. Most existing tools either flatten everything into text or rely purely on OCR. I wanted something that preserves structure and is actually usable in Markdown. Feedback welcome.