Project Red Team uses one evidence engine to produce studies, compare candidate sites, and audit existing studies, with release checks that show what is supported and what still needs review. Each job below states what runs today, what it takes in, what it returns, and where a human analyst makes the call.
The engine aggregates the published regional economic record, runs the project across the ranges of its declared assumptions, and reports where the outcome holds and which variable moves it most.
The engine samples candidate locations along a corridor, re-runs the study logic by candidate, and withholds a final recommendation until the measured ranking is stable.
The engine inventories the quotable claims in a study you did not commission, attaches the evidence found in the document, flags missing support, and runs methodology gates.
Every claim here maps to a generated artifact or a release check. For how these turn into engagements, see services and pricing.