If you're interested please drop an email. I've only worked deeply with pipelines extracting data from documents and I'd be interested in hearing what the challenges with databases are.