Staff GTM data scientist in the San Francisco Bay Area.
I work out where marketing and product money actually pays off, then build the measurement so the answer holds up when someone pushes on it. About eight years across attribution, experimentation and forecasting, now as the founding data scientist at Agaro AI.
Regression-based marketing mix models, multi-touch attribution, A/B and holdout testing, propensity matching, difference-in-differences, funnel and cohort analysis, LTV and CAC modeling, churn and revenue forecasting. Lately: Claude Code routines and subagents, Claude skills in Slack, and supervised-autonomy agent workflows.
Herat University
Stanford Continuing Studies, 2019
Synthetic AI agents that test web pages, so low-traffic teams can still run experiments. You can try the early version.