A/B Testing & Experimentation Framework

A complete experimentation framework for product teams and AI agents. Covers the full lifecycle: structured hypothesis formation, sample size calculation with quick-reference tables, test duration estimation, experiment brief templates, both frequentist and Bayesian statistical analysis with working Python code, post-experiment reporting, and common pitfall prevention (peeking, multiple comparisons, Simpson's paradox, novelty effects). Includes feature flag integration patterns, experiment velocity scoring, and decision frameworks for ship/kill/iterate. No external API keys needed - works with any feature flag system. Whether you're running your first A/B test or scaling an experimentation program, this skill provides the statistical tools and operational templates to make data-driven product decisions.