Crispus Roshan
Free Lesson

Build a Data Pipeline With Claude Code.

45 min
Nov 5, 2026 8:00 PM

By continuing, you agree to Maven's Terms and Privacy Policy.

What you'll learn

Set up Claude Code for data work

Write a CLAUDE.md and use plan mode, so Claude follows your conventions and asks before it acts.

Build a Pipeline From Raw API Data to dbt, Live

Watch Claude Code load API data into DuckDB and turn it into a tested dbt model, step by step.

Catch What the AI Gets Wrong Before It Ships

Find the bug with dbt tests and a data contract, then answer "How do you know the AI's code is right?"

Why this topic matters

Claude Code can write a data pipeline in minutes. Proving it works is the part that gets you hired: AI-written SQL can run without errors and still return the wrong numbers. In this session I'll build a pipeline live, from raw API data to a tested dbt model, then catch the mistake the AI ships. You'll leave with the starter repo and a simple way to check AI-written pipelines, at work and in interviews.

You'll learn from

Crispus Roshan

Crispus Roshan

AI Data Engineer @ Meta | Founder @ Stackle | Ex Google, BofA, Kaiser

Kaiser Permanente
Bank of America
Wells Fargo
Meta
Google
See all products from Cris
Get free access