Ying C.

I turn complex PDFs into clean, structured data — tables with merged cells and multi-column layouts, invoices and receipts, scanned documents — with machine-verified accuracy you can audit. Most PDF extraction breaks exactly where it matters: merged-cell ranges, footnote scope, multi-column reading order, symbols like ✓ ⊗ ● ○, and low-quality scans. That's where I focus: I build document-AI pipelines and measure them against the hard cases, so quality is a number you can check, not a claim. What I deliver: • Complex table extraction from PDFs to Excel/CSV — cell-level alignment, with provenance tracing every output value back to its source cell • Invoice and receipt field extraction (vendor, dates, amounts, line items) with automated quality checks, ready for your accounting tools • Document layout analysis and reading order, scanned-document OCR • Extraction pipelines integrated with your stack via official APIs How I work: fixed-scope or hourly; short diagnostic on your real samples first; you see extraction results on your own documents before committing to a full build. Send me a sample PDF and the output format you need — I'll tell you straight what's easy, what's hard, and what it takes.