Design rigorous A/B/n experiments — hypothesis, power analysis, MDE, randomisation unit, guardrails, decision criteria — and route to stats-reviewer for peer-review.
Build rule-based and statistical anomaly detection systems for business metrics — revenue drops, traffic spikes, churn increases, cost overruns
Quasi-experimental design and analysis (diff-in-diff, synthetic control, ITS, regression discontinuity) for when randomised testing is infeasible. Routes to stats-reviewer.
Design cohort analysis frameworks with SQL queries and visualisation specs for retention, revenue, and churn
Auto-generate comprehensive data dictionaries from database schemas, CSV files, or API responses with column definitions, relationships, and Mermaid ERD
Design ETL/ELT pipeline architectures with data flow diagrams and transformation specs for Supabase and BigQuery
Profile datasets and audit data quality across six dimensions, producing prioritised cleaning recommendations
Analyse A/B test results — significance, CIs, segment cuts, novelty/primacy check, SRM, decision matrix application, and follow-up experiments.