Portfolio
Real problems, real data, real impact. Each case study follows the journey from observation to hypothesis to measurable outcome.
Built a 2-way ML coach recommendation system that outperformed rule-based matching by 80% across 36 countries.
Mapped the booking funnel, identified drop-off points, and ran cohort experiments that nearly doubled conversion.
Found hidden cost leakage in provider payouts and non-completed sessions, recovering ~10% of operational costs.
Built ETL pipelines, flat table architecture, and internal AI tools that made teams self-sufficient in analytics.
Ran multiple experiments across no-shows, cancellations, and booking flows — each driven by a data-backed hypothesis.