Section oneWhat a model is.
A living model, not a report. Each one reconstructs the expectations already sitting inside the share price — the growth, the margins and the returns on capital the market is paying for — from the company's filings and from market data, with base, upside and downside cases built against them.
Every figure that carries weight resolves to its source, and judgements are labelled as judgements, so you can see precisely where a judgement was made rather than a fact recorded — and therefore precisely where there is something to argue about. Then you argue: each model opens in a workbench where you push on any assumption yourself and watch the implied value move against the price, computed by the same engine that built it.
Section twoHow it is built.
Arithmetic is not the model's work. A deterministic financial modelling engine computes every number, so identical inputs produce identical outputs and no figure anywhere is the result of a language model doing mental arithmetic.
The AI orchestrates, and it does that inside skills: narrow, repeatable procedures for one task each, every one with a defined input, a defined output, and a place to record what was judged and on what basis. Before anything is published it has to pass a set of gates:
The engine does the math. The call stays yours.
Section threeThe method.
The first question is not what a stock is worth, but what the market already believes about it. This is the expectations investing method set out by Rappaport and Mauboussin, and it is where each model begins: reverse-engineering the growth, margins and returns implied by today's price. The base is not our view. It is the market's view, reconstructed, and it is not settled until it prices out against the actual market capitalisation over a plausible horizon. The method is old and proven. The medium is new.