The author's claims are fairly loosely worded so it's hard to make much of a determination until the relevant code ships in some form.

For example, the documentation claims both no memory leaks and C/C++ translation. It's not possible to do both in general because C and C++ programs can have memory leaks. Presumably what the language will end up with is the ability to translate some subset of C and some subset of C++ automatically. That may or may not be generally useful depending on the details of the subset.

Even the claim of no memory leaks itself is a bit suspect. GC'd languages don't make that strong a claim. With a sufficiently restrictive language, it's probably possible to achieve but I'm not sure how useful that language would be. My guess is that it'll probably morph into a claim about memory safety which is more reasonable.

Finally, please don't take this as a criticism of the language. It's still too early to have a sense of that. It's perfectly possible for the language to fall some way short of its pitch and be really great. But that's obviously caveated by a variation of the old adage that all of the code I haven't written yet is bug free.