Comment by __mharrison__

Multiple dimensions (more than 2) are hard.

I was at a conference the other day, and my friend (a very smart professor) was asking me if it would be possible to move away from Xarray to Pandas or Polars...

Perhaps using Numba or Cython (with loops) might make it fast but less prone to confusion.

Luckily for me, I mostly stay in tabular data (< 3 dimensions).