Earlier quoted context omitted.
Understanding mass in molar terms is necessary to do a lot of chemistry correctly. The mole is effectively a count of the number of molecules kicking about, although the count is large enough that terms like "quadrillion" don't cut it. (One mole is about 600 sextillion molecules). Knowing how many molecules are around lets you actually compute how much stuff can react in a given chemical environment, and other aspect…
trying to use kg instead of moles when calculating a chemical reaction would be like trying to set up a speed dating event by taking the weights of all the men and women instead of matching them up pairwise. Sometimes it is much easier to calculate based on scalar quantity than it is to calculate based on mass.
On the other hand it is fundamental-ish constant, because it is defined as arbitrarily scaled result of inherently uncertain measurement. And the new definition of kilogram had significantly increased the attainable certainity of such measurement (to the extent that it is uncertain due to practical issues, not by definition). The other proposed replicable definition of kilogram (ie. mass of Si monocrystal wih particular geometry) would fix the definition of mole as some known and defined number, but would be significantly harder to replicate (because it would define how to produce an artifact in contrast to how to directly measure the mass as the ratified definition does)