PDF document generation library from templates

Our company issues dozens of invoices, contracts and acceptance acts every day, and right now each one is assembled by hand in an office editor. We are looking for a developer to build a reusable PDF generation library that takes a template plus a set of field values and returns a finished, print-ready document. The library will become part of our backend, so clean architecture, predictable behaviour and readable code matter to us as much as the visual result.

Each template must support text fields, tables with automatic row wrapping, page numbering and repeating headers and footers. We also need to place raster images on the page: a company stamp and a scanned signature that sit at fixed coordinates and can be scaled without losing quality. The output has to be pixel-stable, because the same contract generated twice should look identical byte for byte, and it must embed fonts so that Cyrillic renders correctly on any machine that opens the file.

Please propose the language and libraries you would use, describe how templates are stored and versioned, and estimate the effort. We value candidates who think about long documents, memory use and the case when a field is missing or too long for its box. Unit tests covering the tricky layouts are part of the deliverable, not an extra.