I don't want to abandon Rust for Julia

This post is mainly just a story I want to share, which may be useful to indicate where Rust is still lacking. But also sort of a call for advice how to use Rust in my project despite the roadblocks

Last year, I learned Rust for one of my projects and immediately loved it. Now, I'm about to start a new project, for which I will be developing a coupled numerical/machine learning model for precipitation prediction (this family of methods is sometimes called Physics Informed Neural Networks or SciML). The only problem is that this research field is so novel that the methods are implemented only in Julia.

I understand why SciML methods are mainly developed in Julia. The language allows for quick development, can be semantically similar to math equations and has no substantial performance overhead. But it means, that I'll have to develop my model in Julia, which I would prefer not to use.

Because Rust to me seems as a much better choice. It's compiled and encourages static linking, which would make deployment of the model anywhere easier. It has good Python interop (necessary for handling meteorological data). And it has features and supporting software we all love - great compiler messages, strong and static typing, error handling, clippy, rustfmt, rust-analyzer etc. etc.

So I don't want to abandon Rust as it is simply great, but I kinda have to if I don't want to spend my project time on reinventing the wheel... If anyone have any idea how can I solve this issue, I'm listening.