Claude for Data Engineers: Every Tutorial

Claude for Data Engineers · All 10 parts

This page lists every tutorial in the Claude for Data Engineers series: 10 videos that take Claude and Claude Code into real data engineering work. We install Claude Code and start a data project, go through every Claude Code concept, write, debug and speed up SQL, build a PySpark pipeline and a dbt project, connect Claude to a database with MCP, build a data-quality skill, use hooks, sub-agents and parallel agents, review pull requests in GitHub Actions, and finish with an end-to-end project from raw CSV files to a dashboard.

Each part has a video and a written post with the exact prompt to type, the files Claude writes, and the real output of every command. Click a part to read it, or click a step to jump straight to it.

Part 1: Claude Code for Data Engineers: Install, Set Up and Your First Data Project

Watch Part 1 on YouTube

1. Step 1: Where Claude Code lives
2. Step 2: Install Claude Code
3. Step 3: Sign in
4. Step 4: First launch
5. Step 5: Ask Claude what the repo does
6. Step 6: Run /init and create CLAUDE.md
7. Step 7: First script and a real run
8. Step 8: Check the numbers
9. Tips for your own repo

Part 2: Every Claude Code Concept Explained

Watch Part 2 on YouTube

1. Step 1: Permission modes
2. Step 2: Permission rules
3. Step 3: Plan mode for a real data check
4. Step 4: CLAUDE.md and memory
5. Step 5: Slash commands
6. Step 6: Context and /compact
7. Step 7: Checkpoints and /rewind
8. Step 8: Models and effort
9. Step 9: Headless runs with claude -p

Part 3: Write, Debug and Speed Up SQL with Claude

Watch Part 3 on YouTube

1. Step 1: Write SQL in the Claude app
2. Step 2: Run it with Claude Code
3. Step 3: Fix a broken query
4. Step 4: Speed up a slow query
5. Step 5: Window functions
6. The checklist: SQL with Claude

Part 4: Build a PySpark Pipeline with Claude Code

Watch Part 4 on YouTube

1. Step 1: Read the raw data and plan
2. Step 2: Write the code
3. Inside the transforms
4. Step 3: Run it with Spark
5. Step 4: Unit tests with pytest
6. Step 5: Check the Parquet, save the commands
7. The checklist: Spark with Claude

Part 5: Build a dbt Project with Claude Code

Watch Part 5 on YouTube

1. Step 1: Open the project and load the seeds
2. Step 2: Staging models
3. Step 3: Marts and a first result
4. Step 4: Tests that catch bad data
5. Step 5: Fix it and run dbt build
6. Step 6: Docs and CLAUDE.md
7. Tips for dbt with Claude Code

Part 6: Connect Claude to Your Database with MCP

Watch Part 6 on YouTube

1. What MCP is
2. Step 1: A read-only Postgres role
3. Step 2: Test the MCP server first
4. Step 3: Add the server with claude mcp add
5. Step 4: Ask questions about real data
6. Safe MCP for data teams
7. Step 5: Connectors in the Claude app

Part 7: Claude Skills for Data Work: Build a Data-Quality Skill

Watch Part 7 on YouTube

1. What a skill is
2. Step 1: Create the skill with Claude Code
3. Inside SKILL.md
4. Step 2: Claude picks it up by itself
5. Step 3: Call it with /data-quality
6. Step 4: Skills in the Claude app
7. Skills that work

Part 8: Hooks, Sub-agents and Parallel Agents for Data Pipelines

Watch Part 8 on YouTube

1. Three tools, three jobs
2. Step 1: A hook that checks every SQL edit
3. Step 2: The hook catches a real mistake
4. Step 3: A read-only SQL reviewer sub-agent
5. Step 4: Two worktrees for parallel sessions
6. Step 5: Test and merge the parallel work
7. Which one to use when

Part 9: Claude Code in CI/CD: Review Data Pipeline Pull Requests with GitHub Actions

Watch Part 9 on YouTube

1. How Claude Code GitHub Actions works
2. Step 1: Quick setup with /install-github-app
3. Step 2: The workflow file, line by line
4. Step 3: Automation mode with a prompt and claude_args
5. Step 4: The same review, run locally
6. Step 5: Fix the bug and re-run the checks
7. Step 6: Headless claude -p and CI safety

Part 10: End-to-End Data Project with Claude Code: Raw CSV to Dashboard

Watch Part 10 on YouTube

1. Step 1: Plan the project in plan mode
2. Step 2: Profile the raw CSV
3. Step 3: Ingest and model in raw, staging and marts
4. Step 4: Data-quality checks catch a bad row
5. Step 5: Build the HTML dashboard
6. Step 6: One command for the whole pipeline
7. The whole project

Each part has a matching video on the TechBrothersIT YouTube channel, in the "Claude for Data Engineers (Claude Code, SQL, PySpark, dbt)" playlist.