Incin: Rust ML Framework That Catches Tensor Shape Errors at Compile Time

Incin is a new deep learning framework written in Rust that encodes a tensor's shape, data type, device, and gradient state directly into its type system. This design means common errors such as shape mismatches or device conflicts are caught at compile time rather than at runtime. A code example in the announcement shows that attempting to multiply incompatible tensor shapes simply fails to compile. The project's primary goal is to explore how much of the tensor contract a type system can realistically enforce while remaining flexible and developer-friendly. Documentation and a dedicated book are available for developers wishing to explore the framework.