🦆 Batch data pipeline with Airflow, DuckDB, Delta Lake, Trino, MinIO, and Metabase. Full observability and data quality.

📡 Real-time data pipeline with Kafka, Flink, Iceberg, Trino, MinIO, and Superset. Ideal for learning data systems.

An example for analyzing time-event data

AI-powered end-to-end e-commerce analytics platform integrating MySQL, advanced SQL modeling, executive BI dashboards (Power BI), and generative AI reporting to deliver insights on revenue growth, retention, operational efficiency, and customer lifetime value

An end-to-end Data Analytics project that demonstrates data modeling, ETL, SQL analytics, Python reporting, and BI dashboarding using a large sales dataset.

Beauty & skincare supply chain AI platform. Social signal-aware demand forecasting (MAPE 12%), CUPED A/B testing (55% variance reduction), X-Learner uplift (AUUC 0.74), DoWhy causal inference, 5 SQL analytics pipelines, MLflow + Evidently MLOps, SHAP/LIME explainability, 120+ tests.

Pipeline de dados em Python e SQL aplicando conceitos de ETL e ELT, com validação de dados, análises analíticas e dashboard visual.

SQL-based Financial KPI dashboard with optimized Power BI queries.

PostgreSQL data pipeline for NFL play-by-play analytics with full ETL, schema design, and optimized SQL views.

Симулятор рыночного мониторинга с автоматическим сбором данных и SQL-аналитикой (MAX, MIN, AVG). 📈🗄

Automated statistical auditing suite for A/B testing and metric validation. Implements adaptive hypothesis testing (Shapiro-Wilk, T-test, Mann-Whitney U) with automated P-value and Lift calculation for data-driven decisions.

This project aims to create a comprehensive data analysis system based on the Wide World Importers (WWI) database.

End-to-end data pipeline with Python orchestration, PostgreSQL Medallion layers, metadata + invalid-record logging, and CI/CD (pytest + flake8 on GitHub Actions).