I would like to get insight into the quality of data my data pipeline is processing. Something like logging metadata about file names created, sizes, rows/columns, etc. I think this could be useful to track down potential issues.