So you've probably learned that if u is an eigenvector, then multiplying u by any scalar gives you another eigenvector with the same eigenvalue. That means that the set of all a*u where a is any scalar forms a 1-dimensional space (a line if this is a real vector space). This is an eigenspace of dimension one. The full definition of an eigenspace is as the set of all eigenvectors of a given eigenvalue. Now, if an eigenvalue has multiple independent eigenvectors, then the set of all eigenvectors for that eigenvalue is is still a linear space, but of dimension more than one. So for a real vector space, if an eigenvalue has two sets of independent eigenvectors, its eigenspace will be a 2-dimensional plane.
I "learned" about them for quantum computing (I think that's mostly linear algebra). I was kind of disappointed they're just vectors I somehow expected them to do something weird (based off the name).
My experience with eigenstuff has been kind of a slow burn. At first it feels like "that's it?", then you do a bunch of tedious calculations that just kind of suck to do... But as you keep going they keep popping up in ways that lead to some really nice results in my opinion.