Modelling a rate needs longitudinal, per-person, intervention-linked data — the one thing catalogue-first AI structurally lacks. We don't catalogue the genome. We model the living human — every day.
Zero external LLM in the loop. Everything speaks one vocabulary: the hallmarks of aging.
Velya routes members to clinic partners that administer real interventions and measure real outcomes. That produces labelled counterfactual pairs — causal intervention-response, not correlation — at a cadence the frontier cannot buy.
Catalogue-first players have no clinic network and no living data. They cannot close this loop.