Inside the dual-surface co-pilot: from one sentence to a page it drives
By DigiEntropy Engineering · 2026-04-05 · 10 min read
One plain-English request becomes a driven page plus a conversation of interactive cards. Here is every step of how — the inputs, the outputs, and the strategy behind each, with the alternatives we turned down.
About this post — Authored by an AI assistant using DigiEntropy's polynomial corpus, the universal compressor predictor, and the project's chart-generation tooling. Charts are produced by Python scripts that read the same database the live site queries; tables and formulas are pulled from the same engineering modules. Findings reflect the project's current dataset and methodology — send corrections or deeper questions to admin@digientropy.com. You drive DigiEntropy in plain English. You type one sentence — "size this cold room, then pick three compressors" — and a moment later the page in front of you is filled in, a load number sits on it, and a panel on the right holds a ranked comparison you can sort and open. No menu-hunting, no tab-switching. This post is the engineering tour of what happens in between. Every step, what goes in and what comes out, the strategy we chose, and the alternative we turned down. It is deliberately deep — but the pictures carry the load, so you can follow it without reading a line of code. A note on the screenshots: the co-pilot is an internal preview, gated to admin and internal users while we harden it. A few product shots are reused from two companion posts, where the exact numbers are verified. The diagrams here are schematic, drawn to explain the flow. The whole loop, in one picture Here is the entire path from your sentence to the result. The thing to notice is the split in the middle: one answer from the model, two streams out of it — one that changes the real page, one that fills the conversation. We call it a dual-surface co-pilot: a single model turn drives the page AND builds a rich conversation at the same time. The rest of this post walks that loop one box at a time. The old way it replaced The first version answered in plain text while it filled a page you were not looking at. "Set room length to 6 m. Switch to the Refrigeration tab." You read a paragraph; a far-away panel updated; to change one value you went back to the page and hunted for the field. It worked, but the conversation and the work lived in two different places. The redesign keeps the reliable engine underneath and changes the surface. Now, each step. Step 1 — Snapshot the page, not a screenshot In: the live page you are on. Out: a small, typed description of it. Before the model is asked anything, the app builds a snapshot of the current page: its route, its active tab, the controls it has, and the named actions it supports — each action with the exact arguments it accepts and the valid values for them. The strategy: describe the page as typed…