Show HN: cuTile Rust: Safe, data-race-free GPU kernels in Rust

cuTile Rust is a new framework that enables writing memory-safe, data-race-free GPU kernels in Rust by extending Rust's ownership model across GPU launches, where mutable tensors are partitioned into disjoint pieces and immutable tensors are shared. The system uses a macro-based approach to JIT-compile kernels through CUDA Tile IR into GPU code, supporting synchronous launches, asynchronous pipelines, and CUDA graph replay. The project is in early research stage and requires NVIDIA GPUs with compute capability sm_80 or higher, CUDA 13.3, and Rust 1.89+.