CUDA programming consists of making a kernel parameterized by its thread id which is used to slightly alter its behavior while it executes on many hundreds of GPU cores; it's very different than general purpose programming. Memory and branching behave differently there. I'd say at best, it will be like traditional programs and libraries with multiple incompatible event loops.

I write CUDA code... I'm aware of it's execution model. In this context, having a rust program act as a kernel to receive inputs and send outputs is still awesome.