So for labeling the documents, a label printer and QR code generator. This is the easy bit. Maybe pasting the label on the back of each document?
For reading the label a small Raspberry Pi with its camera? If we want it to be cutesy, maybe even it’s all integrated into some kind of desk paper-clipping mechanism? So the document gets anchored to the reading space on the desk and then the QR code is read at the same time?
Finally, the medium difficulty bit, some kind of computer orchestration layer that takes the RPI output and actually sets up the workspace?