It depends on what you mean by application, self-evolving/learning programs is basically impossible at this point because of the extremely large search space of

It depends on what you mean by application, self-evolving/learning programs is basically impossible at this point because of the extremely large search space of possible code within the syntax (how many possible programs exist within say 10 lines, and of those how many would actually run without errors and produce something useful?), not to mention it would produce unmaintainable code in the first place. But what is useful is giving the language users the ability to write their own domain specific compilers, for example extending the syntax to represent the math more closely making it easier for non programmers and for self-documenting code, doing symbolic analysis to simplify the code and finding the best way to solve it, exporting directly to other formats like LaTeX and even extending the compiler to automatically generate and compile the optimized gradient code for any function you write.