Topic
AI
Models, evaluation and what it means to think alongside machines.
AI
What a small model knows that a large one forgets
On scale, memory and why a smaller model can be easier to reason about — and what that means for the tools we build.
AI
Evaluations are the product
If you cannot say what good looks like, you cannot ship it. A sketch of how evaluation shapes everything downstream.
AI
Reading attention maps like margin notes
A way of looking at model internals that borrows from how we annotate books: lightly, and with questions.