Traid technical blogs
Technical writing on the platform: how assistants think, how apps are built and kept safe, and what we're working on.
We made the AI write in a language it can't break out of
Our assistants build apps you can run in the browser. Getting there meant giving the model a language of its own.
How a Traid assistant works
A technical look at how Traid assistants work: two kinds of agents, on-demand generation of a Clojure-based DSL, and why interpreting that DSL in a sandbox is safer than running AI-generated code directly.