dbt for Analytics Engineers
Everything an analytics engineer needs to learn, ship, and interview with dbt — from your first model to a production-grade dbt Cloud project.
dbt — short for data build tool — is the transformation layer of the modern data stack. Whether you're at a 10-person startup running dbt Core against BigQuery or a Fortune 500 team using dbt Cloud against Snowflake, the workflow is the same: select from raw source data, transform it through layered models, test the output, and ship to production through Git.
Analytics engineers own this layer. If you can build, test, and document dbt models, you can hold the entire transformation stack of most companies on your own. That's why dbt has become one of the most in-demand skills in data hiring — and why every analytics engineering role posted in the last two years asks for it by name.
This hub is the curated path to learning dbt the way it's actually used on the job. The articles below cover the framework end-to-end. The capstone is a complete dbt Cloud + BigQuery + Looker build you can ship to GitHub and discuss in interviews.
By the end of this path you can…
- Set up dbt Core locally and dbt Cloud in the browser
- Structure a dbt project the way production teams structure them
- Write idiomatic dbt models with refs, sources, and macros
- Add data tests, snapshots, and documentation
- Deploy dbt with GitHub and dbt Cloud jobs
- Answer the dbt interview questions hiring managers actually ask
- Ship a portfolio-ready dbt project on BigQuery
From beginner to job-ready.
- 01 · FoundationsWhat dbt is, how it fits in the modern data stack, and the difference between dbt Core and dbt Cloud. Open this guide →
- 02 · Project structureSources, staging, intermediate, and mart layers — the layered modeling pattern every production dbt project uses. Open this guide →
- 03 · ModelsWriting models with refs, materializations (view, table, incremental, ephemeral), and configs. Open this guide →
- 04 · Tests & documentationData tests (unique, not_null, accepted_values, relationships), generic tests, and dbt docs. Open this guide →
- 05 · Snapshots, macros, JinjaTrack slowly-changing dimensions, DRY out your SQL with macros, and template with Jinja. Open this guide →
- 06 · Deploy + CIdbt Cloud jobs, CI in GitHub, and the deployment workflow real teams ship with. Open this guide →
- 07 · CapstoneThe full BigQuery + dbt Cloud + Looker capstone — a portfolio piece you can talk through in interviews. Open this guide →
Read the playbook.
- dbt
dbt Tutorial for Beginners: Learn dbt, the Data Build Tool
A fast, practical dbt tutorial for beginners. Go from setup to your first dbt project with real SQL, ref(), tests, docs, snapshots, and incremental models.
- dbt
dbt Project Structure: Staging, Intermediate, and Marts Done Right
A proven dbt project structure you can copy today: clear folders, schema mapping in dbt_project.yml, and tight naming conventions. Includes SQL, YAML, and real-world tips.
- dbt
dbt ref() vs source(): Model-to-Table References Explained
Use ref() for dbt-built models and source() for external raw tables you declare in YAML. This guide shows exact SQL/YAML, selectors, and pitfalls so your builds run in the right order.
- dbt
dbt Tests: dbt test, Data Test, and Unit Test Guide
A practitioner’s guide to dbt tests. Learn generic (schema), singular, and custom tests; how to run dbt tests; what happens when a test fails; and how to scale, package, and govern them.
- dbt
dbt Snapshots: dbt snapshot Playbook (dbt developer hub)
A practitioner’s guide to dbt snapshots: configuration, timestamp vs check strategy, performance, deletes, and testing—no SCD theory rehash.
- dbt
dbt Semantic Layer for Analytics Engineers: MetricFlow & dbt Labs
A deep, practitioner-first guide to the dbt semantic layer: define metrics once, serve them everywhere. We cover MetricFlow, security, integrations, and real query patterns.
- dbt
dbt Macros & Jinja Tips Every Analytics Engineer Should Know: Expert Guide
Learn how dbt macros and Jinja can transform repetitive SQL tasks into dynamic, reusable code, enhancing scalability and efficiency in data projects.
- Fundamentals
How Analytics Engineers Can Implement Incremental Models
Learn how to implement incremental models in dbt, optimizing data processing by handling only new or modified data. Explore types, benefits, and examples.
- dbt
dbt Cloud vs Core: Feature Comparison 2025—Comprehensive Guide
Compare dbt Cloud and Core to understand their features, costs, and operational differences. This guide helps data teams make informed decisions.
- Interviews
30 dbt Interview Questions and Answers (From Real Screens)
Get 30 practitioner-grade dbt interview questions and answers grouped by fundamentals, modeling, tests/deployment, and scenarios—with real SQL/Jinja examples.
Show, don't just claim.
- intermediate · open →
Data Forge: The Lost Metrics
Metric-layer recovery and analytics debugging project (dbt + BigQuery)
- intermediate · open →
Sports Equipment Pro Shop
E-commerce orders, inventory, and revenue modeling project
- beginner · open →
Champion Fantasy League
Event and performance analytics modeling project
Common questions about this topic.
Should I learn dbt Core or dbt Cloud first?
Start with dbt Cloud if you're new — it removes the local-environment setup so you can focus on modeling. Move to dbt Core once you're comfortable; you'll need it for production deployments at most companies. Knowing both is the right answer for interviews.
Do I need to know SQL before learning dbt?
Yes. dbt is a SQL framework — it doesn't replace SQL, it organizes and tests it. If you can write window functions, CTEs, and joins, you can write dbt. If you can't yet, work through the SQL practice topics first.
Is dbt worth learning in 2026?
dbt is the de-facto transformation tool in analytics engineering. Nearly every analytics engineering job posting lists it. dbt Labs continues to ship major features, and the ecosystem (Coalesce, SQLMesh) reinforces — not replaces — the patterns dbt taught the industry.
What warehouse should I use for learning?
BigQuery has a generous free tier, integrates cleanly with dbt Cloud, and is what we use in the portfolio capstone. Snowflake and Databricks work fine too — the dbt project structure is identical across warehouses.
How long does it take to learn dbt well enough to use at work?
Two to four weeks of consistent practice for the fundamentals (sources, models, tests, docs). Two to three months to be comfortable with macros, Jinja, incremental models, and dbt Cloud deployments. The capstone project takes most people 20–40 hours.
What are the most common dbt interview questions?
Materializations and when to use each, incremental model strategies, slim CI vs full CI, how to test for referential integrity, how to handle slowly-changing dimensions with snapshots, and how to structure a dbt project (sources → staging → intermediate → marts).
Adjacent topics.
SQL for Analytics Engineering
The query language dbt is built on.
Data Modeling
Star schemas, normalization, and the modeling patterns dbt projects use.
BigQuery
The warehouse most dbt learners run against — including our capstone.
Interview Prep
What hiring managers actually ask for analytics engineering roles.
Start practicing this topic.
Graded exercises with hints, worked solutions, and a GPT tutor. Free to start, no credit card.