Create dated dataset files for data scientists before daily jobs

Your cloud data drops arrive untracked and cause delayed model runs and messy lineage. Data scientists get timestamped dataset files and references so training runs before the next daily job.