
I Built a Tested Data Pipeline for Dublin Bikes, for €0
How I collect every Dublinbikes station every 15 minutes, model it with dbt and DuckDB, test it daily and feed a live dashboard, using only free tiers.
Tag archive

How I collect every Dublinbikes station every 15 minutes, model it with dbt and DuckDB, test it daily and feed a live dashboard, using only free tiers.

Key takeaways Agentic data engineering uses autonomous AI agents to build and maintain data...
Fivetran closed its dbt Labs merger and announced dbt Core v2.0 under Apache 2.0. We compare the claims with dbt's licensing docs and what to do next.

If you've worked in a dbt project for more than a few weeks, you've probably hit this moment: you...

dbt built the modern analytics workflow. Now the transformation layer is moving upstream — and the...

In this article I'll show you how a dbt incremental model that looked perfectly reasonable wiped out...
I built a small warehouse, fed it a batch of orders with twelve defects planted in it, and watched...
Many dbt projects start with the same two tests on every key: unique and not_null. They are cheap,...
How saved KPI totals and a simpler reference calculation expose reporting errors that a full...

Two connectors and a sources.yml is the easy half. The two systems share no identifier, so the real question is what fraction of your customers actually match — and what you do with the rest.

There is a bug in analytics SQL that never throws an error. select o.customer_id, sum(o.amount) as...

datanika-mcp is on PyPI and the official MCP registry: 25 tools that let Claude, Cursor, or any MCP client browse connections, preview tables, compile dbt models, and read run logs. Read-only by default.