Read data from a (text based) PDF

0

I am trying (without any success) to read data (ie. employee id) from an existing text-based PDF (test pdf created via MS Word).
The documentation is explaining the different actions (ie. wait for ocr ai, wait for text-layer), but not how they actually work together or how to use them.

Maybe I am completly wrong in trying to use "OCR AI recognition", "Wait for OCR AI" etc.....

My goal is:
* A configured hot-folder puts a pdf into a workflow (=> working)
* A step/action/etc. reads the employee id from the pdf and puts it into the wf instance into "EmployeeID"

I also looked into OCR AI projects (Systems Settings)... but no clue how this should work. I found no information how to create a dll or if a custom dll is needed at all.

Any ideas?