Very interesting to use Haxe for the core interpreter. I compile my own language into a string of tokens that can be interpreted by a simple stack-based VM (which also enables compiling other languages like a sunset of Yarn Spinner), but in my system I need to create an interpreter myself for every language I want to target. Not difficult (the VM is trivial) but a chore. I need to look into how this works with Haxe, thanks for the link?