Interactive Explainers
Study pages for engineers who need to understand a mechanism well enough to make a call. Each section is one claim, proved by an interactive figure you can push to its flip point; every sentence traces to a source.
How Jev works: calibrated decision models
Jev promises decisions with honest probabilities in milliseconds. How does it work, and how close does a stock open model get?
AI API Pricing
What do AI models cost through an API, how have prices changed since 2023, and how do you work out which model is cheapest for you?
GRPO: Group Relative Policy Optimization
How GRPO turns a handful of sampled answers into a training signal without a value network: a primer, four interactive studies and the decisions they settle
MCP 2026-07-28: Stateless Core, Extensions, Tasks
What the 2026-07-28 revision of the Model Context Protocol changes and how the new mechanisms behave: seven stepped figures, their notes quoted or paraphrased from the specification and the extension specs, plus what to build with today