Visual Integrity Technologies builds the extraction layer between PDF files and the systems that need to understand them: AI models, enterprise platforms, engineering workflows, and applications.

Our core technology is a C++ PDF engine three decades in the making. It converts PDFs into 16 formats including DWG, DXF, SVG, PNG, TIFF, and EPS. It creates PDFs from scratch, merges, stamps, and manipulates pages, and extracts text, structure, and meaning from “live” PDFs (PDFs with native vector geometry). Built across 30 years by solving the hardest problems in PDF: malformed files, vendor quirks, spec violations, coordinate edge cases.

The moment has arrived to expose what's in the PDF to AI pipelines as structured, machine-readable JSON. Every vector object, coordinate, layer, and attribute ready for models and pipelines that can now reason over them. Not pixels. Not inference. The actual geometry of the document, queryable and precise.

A “live” PDF is not a picture of a document. It's everything the architect or engineer put into it. Every seat in an aircraft cabin, every wire run in a schematic, every room in a floor plan is a queryable data point with exact coordinates. If a picture is worth a thousand words, a live PDF is worth 300,000 lines of JSON.

Founded in 1993. We serve aerospace, engineering, architecture, construction, fintech, legal, energy, and manufacturing enterprises where accuracy is mission-critical.

Products: PDF Intelligence API · pdf2cad · PDF FLY · ConvertPDF.online · Developer SDKs 

Co-founded by Jan Homan, Voorhout, Netherlands and Jean Haney, Portland, Maine.

Visual Integrity Technologies LLC

Org chart

Teams

Offices