evidence-dev

A monorepo combining data modelling in dbt with data viz using Evidence

A Meltano extension for Evidence.dev, built with the Meltano EDK

NOAA snowfall data analytics pipeline

A public-data warehouse built end to end: dlt → DuckDB → dbt → Polars → Evidence, orchestrated by Dagster. Rebuilt from live sources on every push; the warehouse itself is published monthly.

Data about food safety ratings from The Danish Veterinary and Food Administration

This project is an end-to-end ETL solution for processing and analyzing US flight data with Apache Spark and Iceberg. It features pipeline unit tests, data quality checks, and an interactive dashboard. Explore it below! ↓

End-to-end BI solution for the Sakila database. Includes an automated ETL pipeline with Python & DuckDB and an interactive KPI dashboard in Evidence.dev.

Personal media collection dashboard — Evidence.dev static site built on top of personal-warehouse dbt marts (BigQuery). Deployed via Netlify, refreshed daily.

Projeto de Business Intelligence e Data Analytics baseado em Modern Data Stack com Evidence.dev e DuckDB para monitoramento de métricas escolares e operacionais.

End-to-end data pipeline and analytics dashboard for Airbnb SP. Uses a Medallion Architecture on AWS (S3 + Athena), dbt for dimensional modeling, and Evidence.dev + DuckDB for high-performance static reports

Containerized, reproducible retail analytics pipeline: MySQL -> Python ETL -> PostgreSQL warehouse -> dbt star schema -> SARIMA/Gradient-Boosting demand forecasting + promo OLS A/B experiment -> Evidence.dev dashboard, orchestrated by Apache Airflow. One docker compose up.

dashboard prezență și rezultate alegeri

Agency marketplace is main skill sets of people and workflows in a BAU agency.

Projet de visualisation et d’analytique avec les données ouvertes du gouvernement du Québec

A complete, running analytics stack: model2data -> dlt -> DuckDB -> dbt -> Evidence. Clone it and run it with no account, no credentials and no cloud project; point it at a real database by changing one connection string.

Where a wholesaler's win-back budget should go: 352 At-risk customers, not the 1,556 Hibernating ones. 1.07M invoice lines, DuckDB to dbt to a live Evidence dashboard.