Follow the build

Build log

A running, dated record of what shipped, what it cost to run, and what comes next. Entries are written the night they happen and stay as written. Dates are ISO. Costs are measured, not estimated.

The log

Three entries. One of them is this page.

The product is not available yet. What is available is this log, the agent on the home page, and a list that hears first when either changes.

2026-08-23entry 001 · shipped

Day zero: the site you're reading, and a resident agent you can talk to.

Tonight the site went live with a working agent on the front page. It knows one document, it cannot do anything in the world, and every answer it gives is metered and visible. Below is how each piece is put together.

The site. The whole thing is one Cloudflare Worker serving static assets, with no build step. That is the same shape as the product. It was built in a single overnight session by an AI coding agent working from the product overview and a walkthrough of the live board.

The agent. The window on the home page is Jeffery, this deployment's resident agent; every deployment names its own. He runs on Workers AI, on Llama 3.3 70B instruct in its fp8 fast form. His memory is exactly one document, the product overview, and nothing else. He cannot act: no tools, no tray, no email, no calendar. He runs on a separate, read-only deployment, not the team's own board.

How a question is answered

The overview was chunked the way the product chunks: 1,100 characters with 150 characters of overlap, which came to 89 chunks. Each chunk was embedded with bge-m3 on Workers AI and upserted into Vectorize, and mirrored into a D1 FTS5 table. Every question runs two legs in parallel, a vector search against Vectorize and a BM25 match against the full-text table, and the two lists are fused with reciprocal rank fusion before the model sees anything. The product runs four legs over a whole company's memory; a single document needs these two.

Voice in, voice out

Voice in is the browser's own speech recognition, with Whisper large v3 turbo on Workers AI as the fallback when the browser has none. Voice out is the Aura-1 text-to-speech model on Workers AI. Both are optional; typing works the same.

The graph behind every page. It is a hand-written WebGL point cloud, about 2,600 nodes on desktop, coloured with the same hue grammar the real board uses for entity kinds. Newcomers streak in as meteors. A focus channel lets the agent steer it after every answer, the way the product's show_on_graph tool steers the real constellation. It is decorative: the page reads fine without it.

The list. The early-access form writes to a D1 table and sends one email through Cloudflare Email Sending to one inbox at Duval Software. Turnstile guards the form.

The meter. Every model call the demo makes is recorded in a ledger table with its surface, model and size, the same way the product meters its own calls. The demo carries daily caps of 800 chat turns and 300 voice replies; when a cap is reached the demo says so and points to the list.

What one turn costs. Measured tonight: about 1,700 prompt tokens and roughly 45 neurons to the first token. That is fractions of a cent per answer, which is the point of running it on Workers AI.

first measured turn · ~45 neurons · < $0.01
2026-08-23entry 002 · shipped

The overview, verified against the code.

The document Jeffery remembers was written and checked against the repository today. It covers what the system is, the top features, the board with its windows, dock and apps, the memory system, the graph and the avatar, the resident agent and the fleet, every channel it talks on and every source it listens to, approvals and the writeback rail, the two MCP doors, the Hand, the architecture on Cloudflare, and what it costs to run. It is the only thing the agent on the home page knows, and it is the source for everything on How it works.

upcomingno date yet

Next.

The plan is to build the rest of this where people can watch. Nothing below is promised; it is what the team intends to put on this page.

  • Recordings of the Shipwright building a feature end to end: plan, build, a machine gate in a sandbox, pull request to staging, CI, browser test judged by a different model lineage, and the merge request left for a person.
  • A worker being born in the Foundry, with every step streamed to the Hive as it happens.
  • A machine paired through the Hand, from the single pasted line to the first command.
  • The face first speaking: the constellation rearranged into a head, the jaw tracking live audio, and the detonation back to the cloud.
  • A weekly cadence: a changelog entry here, a demo you can touch, and a ledger of what that week's model calls cost.

Each of those becomes an entry above, dated the day it ships, with its cost chip.

Early access

Hear about each entry first.

The product is not available yet. When a new entry lands here, the early-access list hears before anyone else. One address, one inbox, nothing else.

Join the early-access list