Writing
Each essay takes one concept and explains it completely — not as a tutorial, not as a summary, but the kind of explanation you return to when you want to genuinely understand something. Built from first principles, with original diagrams.
How agents remember, plan, use tools, and coordinate. The internals of agent cognition explained from first principles, with original diagrams.
Transformer architecture, attention mechanisms, embeddings, and training dynamics. A systems engineer's view of how language models actually work.
Bayesian inference, entropy, gradient descent, measure theory. The mathematical structures that make modern AI possible, derived from scratch.
Embodied AI, robotics perception, sensorimotor learning, and world models. What it takes for a machine to understand and act in the physical world.