Modern Data Stack: Warehousing, dbt & Airflow
Welcome, champion! 🔥
You've mastered 0 of 169 lessons. Start your first lesson free — no signup needed.
About this course
Modern Data Stack: Warehousing, dbt & Airflow teaches you to build analytics pipelines on a cloud data warehouse with dbt for transformation and Airflow for orchestration. The first lesson traces one metric, daily revenue, from a raw orders table through a dbt staging model to a reporting table, which is the extract, load and transform flow the whole course builds on. From there the course covers ELT versus ETL, warehouses, lakes and lakehouses, columnar storage and how warehouses bill compute, dimensional modelling with facts, dimensions, grain and slowly changing dimensions, ingestion with CDC and managed connectors, dbt sources, refs, materializations, incremental models, Jinja macros, tests, snapshots, documentation and the semantic layer, data quality and contracts, Airflow 3 DAGs, TaskFlow, assets, sensors and deferrable operators, orchestrating dbt with Airflow, warehouse performance and cost, governance and CI for analytics code. The course has 169 lessons and about 1127 minutes of content. Each lesson is a short narrated reading card followed by MCQ practice with an explanation for every answer.
How the lessons work
Each lesson starts from a concrete artefact: a few rows of raw data, a dbt model in SQL, a schema.yml with tests, or an Airflow DAG in Python. The idea is explained against that example, then MCQ practice asks what a model produces, which materialization fits, why a test fails or when a DAG will run, with an explanation for every answer. dbt examples use dbt Core 1.x and Airflow examples use Airflow 3 with TaskFlow and assets, so you learn the current tools rather than legacy patterns.
Who this course is for
Data analysts who write SQL and want to build reliable pipelines, aspiring analytics engineers and data engineers, and students and freshers preparing for roles that list dbt, Airflow, Snowflake, BigQuery or data warehousing. You should know SQL well; the course teaches the modelling, testing and orchestration around it. If your SQL needs work first, take SQL for Analysts; for Spark-based processing, Data Engineering: ETL & PySpark is the companion course.
MCQ practice vs video courses
Video courses on dbt and Airflow usually build one demo project from start to finish. The problems you meet at work are smaller and harder to see: an incremental model that silently skips late rows, a fact table at the wrong grain, a DAG that backfills months of runs on deploy, or a warehouse bill that doubles after one change. This course puts you in those situations with a few lines of SQL or Python and asks what happens, then explains why.
Best way to learn the modern data stack
Set up a small dbt project on a free-tier warehouse or DuckDB and load one sample dataset, such as orders and customers. After each lesson, add the model, test or snapshot it describes to that project. Answer each MCQ before reading its explanation. When you reach the Airflow topics, schedule your dbt project from a local Airflow 3 instance, which ties most of the course together in one working pipeline.