If anyone is working with US healthcare claims data you might find The Tuva Project interesting. I ran across it yesterday. They are using dbt to transform the most common file layouts into data marts. https://thetuvaproject.com/

Wow - that looks really cool @BuzzCutNorman! Healthcare data is hard to manipulate, in part because there isn't good prior art, all the solutions are proprietary and closed sources. Also, not having good 'publicly available/shareable/demoable' datasets makes it hard to develop something open like this. Thanks for sharing!

@aaronsteers that is all so very true. In the realm of claims data there are the Medicare Claims Synthetic Public User Files (SynPUFs). https://www.cms.gov/Research-Statistics-Data-and-Systems/Downloadable-Public-Use-Files/SynPUFs