ML with Rust and Category Theory (hghalebi.github.io)

A new working draft titled "Category Theory for Tiny ML in Rust" has been announced, aimed at bridging the gap between category theory, Rust programming, and tiny machine-learning systems. This book provides a unique perspective on machine learning, treating it not just as numerical computation but as a structured system of objects, transformations, and constraints. Authors Hamze Ghalebi and Farzad Jafarranmani leverage category theory as a practical engineering tool, where domain objects are defined as Rust types, morphisms are typed transformations, and compositions represent executable structures, ultimately rendering tiny ML systems mathematically concrete. This project is significant for the AI/ML community as it emphasizes the integration of theoretical frameworks with practical implementations, highlighting how abstract concepts can enhance software architecture in production AI systems. By inviting public feedback on the draft—ranging from technical clarity to practical examples—the authors encourage collaborative development, fostering a richer understanding of the intersection between mathematical rigor and operational AI. As the draft evolves, contributors can explore the accompanying code on GitHub and engage in workshops to deepen their grasp of the tiny ML pipeline structured through Rust, making this an essential resource for developers and researchers alike.