// SIX ENGINES
One engine, one question.
A single report runs one engine against one Key Intelligence Question. Six engines, each purpose-built for a different class of question. Four are report products; two are methods that support the other four.
Rolling updates on topics, actors, and regions. Signal, noise, and what actually shifted.
Reference profiles of people, organizations, and systems. The canon you should have had before the question came up.
Scenario forecasting with calibrated confidence, alternative hypotheses, and pre-declared indicators.
Target system analysis, vulnerability mapping, and opportunity-window assessment. Where the leverage is and when it's open.
A method, not a product. Tests claims, weighs evidence, and produces calibrated confidence for the other engines.
Structured extraction at scale. Rosters, timelines, and datasets built to a declared schema — feeding the other engines or standing alone.