Python/OCR developer

we needed for process of scanning paper documents with a cover page, which holds a barcode. All documents should be scanned without OCR recognition and later saved as PDF. Besides of the PDF - the barcode value must be stored in a separate xml file and hold the same file name as the scanned PDF file. A barcode cover page must also act as a job separator - means if a first cover page is scanning, it only must take paper documents for the first job until another cover page is following with additional paper documents - means each pile of documents has to be separated by the barcode on the cover sheet. Sending scanned PDF and XML documents via REST API would be the optimum - but this can be discussed.