I really like Visidata for "exploring" csv-type data.

Its a vi(m) inspired tool.

It also handles xls(x), sqlitedb and a bunch of other random things, and it appears to support parquet via pandas:

https://www.visidata.org/docs/loading/