It's a good all-round primer, well written. Would love to hear more about larger-than-memory tasks and running local Dask clusters. I processed many-a-dataset that way that would normally make pandas choke.