← Selected work 
Own build
Document rebuild pipeline: Word in, structured data, clean PDF out
A workflow that takes a Word document, extracts its content into structured data, and renders a clean .docx and .pdf, verified by re-rendering the output.
The situation
Rebuilding a document by hand loses structure and introduces formatting drift. The goal was a repeatable pipeline: drop a file in a folder, get consistent, machine-readable output.
What we built
- Extraction of the document’s content into a typed data file
- Templated rendering back to .docx with consistent styles
- PDF rendering through LibreOffice, then a re-render check to catch layout breakage
- A folder-based workflow, so a non-technical user only drops a file in
What carries over
The same pipeline shape works for proposals, reports, and any document a team keeps rebuilding by hand.