As I think about this more, the point may be that .NET is most rewarding for developers at either end of the experience spectrum.
As a very experienced developer, it's easy to get into the low-level framework type features like remoting (and building RealProxies), unsafe code (pointer manipulation), asynchronous programming with io completion ports, etc. All of these features allow experienced developers to create from scratch very powerful abstractions.
In turn, inexperienced developers can easily take advantage of these abstractions to build something quickly. It is this group that the author is referring to (and probably belongs to). Once a member of this group needs to do something outside of the realm of existing re-usable components, it takes a more experienced developer to help.
As a very experienced developer, it's easy to get into the low-level framework type features like remoting (and building RealProxies), unsafe code (pointer manipulation), asynchronous programming with io completion ports, etc. All of these features allow experienced developers to create from scratch very powerful abstractions.
In turn, inexperienced developers can easily take advantage of these abstractions to build something quickly. It is this group that the author is referring to (and probably belongs to). Once a member of this group needs to do something outside of the realm of existing re-usable components, it takes a more experienced developer to help.