Category Theory for Tiny ML in Rust — Workshop 4

The Tiny Training Loop July 23, 2026 Last session before the summer break. We restart in September. In this final workshop before summer, we will connect the ideas from the previous sessions into one tiny ML training loop in Rust. We will move step by step: Input -> Logit -> Probability -> Prediction -> Loss -> Gradient -> Updated Model The goal is not to build a huge neural network. The goal is to understand how Rust types, typestate, traits, and category-theory-inspired thinking can make illegal ML transitions harder to write. We will focus on simple live coding, visual explanations, and compiler errors that are actually useful. No need to be an expert in Rust, ML, or category theory. Curiosity is enough. This session will also help shape the next season. After the summer break, I would love the workshops to become more community-driven, with more participant ideas, tiny demos, shared questions, and suggested topics. Come for the tiny ML model.

Stay for the compiler saying “no” before your bug becomes philosophical.