Explainers.

Interactive pieces about systems that are hard to picture—some a real-time animation you can scrub and scroll, some a map you take apart by clicking it.

• 5 experiments · 4 viewers

  ChatGPT Sites and Claude Artifacts

  When the chatbot hosts what you built, what did you get? Open one team board as four different people, save a record and get it back, break a permission, and follow a summary to the account it draws on.

• 4 scenes · 3 cache tiers

  Scion, explained through Git

  Anthropic replaced Git for its monorepo. Follow one edit through both systems: what moved from the clone to a shared service, what a changeset records, and how a job gets its files through three tiers of cache.

• 17 chapters · 7–8 minutes

  Inside the Machine

  How a prompt becomes tokens. One request through a trillion-parameter mixture-of-experts model on a rack of seventy-two accelerators: across the network, through a router that wakes sixteen experts out of eight hundred and ninety-six, and back out as streamed text.

• 5 levels · Click to zoom

  Anatomy of a B300

  A top-down map of one NVIDIA accelerator that opens whatever you click. Down through the package, a compute die, a cluster, a single multiprocessor, and a stack of memory twelve dies tall—until you reach the registers, where the arithmetic actually happens.

• 5 chapters · 7 instruments

  Cloudflare, in pieces

  A guide to requests, data, shared state, and background work. Send a request and watch the cache answer it, read a changed value from two places, run a Durable Object room, and deliver a queued message twice.