Writing
How AI systems work in practice.
Long-form, visual explanations for builders who want intuition, implementation detail, and honest lessons from production.
First series in progress
I’m writing the first article now.
It’s about agent computers and sandboxes. I’m building the diagrams and examples now. Subscribe if you’d like it in your inbox when it’s ready.
01Agent computers and sandboxes
02Background and always-on agents
03Attention without the wall of math
04Production evals that catch real failures
05Fine-tuning, LoRA, RLVR, and GRPO