Tributary
A data science case study on a simulated lead marketplace: business questions turned into decisions with identity modeling, calibrated models, and an honestly read experiment, built by directing AI agents under an auditable harness.
Tributary is a data science case study. A simulated two-sided marketplace sells personal-loan leads through a six-tier waterfall auction, and its data arrives the way real marketplace data does: in three systems with no shared key. The case study asks the questions the business would ask — which channels earn their spend, what a duplicate consumer costs, where auction floors should sit, who a nurture campaign should reach — and answers them with identity modeling scored against hidden ground truth, calibrated models, and an experiment read at its actual power. It was directed by one person and built with AI agents, and the direction itself is kept as an auditable record.