I'm developing a memory safety model that dramatically improves upon Rust. You can read about an early iteration of my model on Verdagon's blog.

I am aiming for this model (or something like it) to be adopted by a Rust successor language such as Mojo or Carbon. I am making good progress: the Carbon designers were inspired by the blog post, and they intend to demonstrate my core thesis: that certain patterns of mutable aliasing are safe, and we only need to identify and restrict the unsafe patterns.

I am currently doing this work as a weekend "hobby", because there are virtually no opportunities to get paid to develop memory safety models. You can help change this! With financial support, I will be able to dedicate more of my week to working on the future of memory safety.