Category Theory for Tiny ML in Rust — Public Workshop

Modern AI frameworks made machine learning accessible.
But accessibility is not always understanding.
In this workshop, we will explore the ideas behind Category Theory for Tiny ML in Rust, a public working draft and Rust-based learning project for engineers who want to understand AI systems below the framework layer.
The goal is not to replace Python.
The goal is not to use category theory as decoration.
The goal is to rebuild tiny ML concepts from first principles using:
Rust types
typed transformations
composition
training loops
category theory as an engineering tool
We will look at how a small ML pipeline can be understood as a sequence of explicit transformations:
Text → Tokens → Training Pairs → Model State → Prediction → Loss → Updated Model State
The workshop will be held in Paris and streamed online for remote participants.
It will not be recorded.
This is intentional: the session is designed as a live, interactive discussion where participants can ask questions, challenge the ideas, and give feedback on the public draft while it is still evolving.
Who it is for:
ML engineers who want to understand what frameworks hide
Rust developers curious about AI
systems engineers interested in typed design
category-theory-curious engineers
technical founders and builders who prefer first-principles learning
You do not need to be a category theory expert.
You do not need to be a Rust expert.
You only need curiosity about how tiny AI systems can become more explicit, composable, and understandable.
Not abstraction cosplay.
Executable structure.