Hello,
First, the phrase you asked "(Bananas)".
I'm an experienced Software Engineer, in my work I engineer, develop and maintain several applications, using several operating systems.
To be fair, I've never worked with OCR or PDF files before. But I think the challenge of this project is not reading the PDF and applying and OCR library. The challenge will be do it quickly enough!
I'm assuming that the PDF isn't text and I'll be using the OCR library to recognize the characters (handmaid, perhaps?).
I don't think the solution is very easy, so I propose two milestones. The first you provide me the pdf files and I try to do a proof of concept.
Then, if successful, we implement the full project scope.
I'm assuming that the scope is the "backend OCR reader/storage".
Best Regards