synthetized dataset based on documentation not publicly available q&a pairs in json format (for now).

pdf file that has to be reconfigured in JSON