experiment-analysis

Design of Experiments and Analysis

Scientific Research Reasoning Engine for analyzing experimental evidence, evaluating scientific claims, and generating structured research reports.

African language Speech Recognition - Speech-to-Text

A hands-on tutorial for understanding experimentation systems by building one

Statistical validity engine that audits A/B experiments across 8 checks:SRM, power, peeking, SUTVA, and more.

Statistical A/B testing analysis of marketing campaigns using hypothesis testing to compare performance based on key business metrics.

PyTorch simulator for Federated Reconstruction (FedRecon) on MovieLens 1M.

A/B testing analysis of mobile-game retention using frequentist, bootstrap, Bayesian, and product-decision methods.

합성 제품 이벤트를 데이터 품질·A/B 테스트·3개 가드레일 검증과 의사결정 메모로 연결한 분석 워크플로

Randomised campaign analysis and doubly robust targeting evaluation using Python, SQL and scikit-learn.

African language Speech Recognition - Speech-to-Text

Marketing A/B testing case study with conversion uplift, hypothesis testing, confidence intervals, effect-size interpretation and power planning.

Ant colony simulation with pygame visualization, configurable experiments, and result analysis.

Product analytics decision engine for evaluating SaaS onboarding experiments using BigQuery, A/B testing, Power BI, and Streamlit.

This repo has a complete demonstration of performing experiment tracking with mlflow

Ecommerce growth analytics platform with funnel intelligence, conversion drop-off analysis, retention analytics, experimentation dashboards, and executive KPI monitoring.

Audit completed A/B experiment readouts for SRM, peeking risk, practical significance, guardrail movement, and pre-period imbalance. Zero dependencies. Structured PASS/WARN/FAIL reports in JSON, Markdown, and HTML.

Release intelligence for AI agents that turns behavior, product metrics, statistical investigation, economics, and human feedback into evidence-backed SHIP/HOLD/ROLLBACK decisions.

Корпус синтетических A/B-кейсов: в каждом ровно одна заложенная ошибка. Материал для проверки AI-аналитика.

FastAPI web app that analyzes A/B test CSVs and generates AI-driven experiment insights with automatic stats, uplift, and p-values.