Why Git for data work

What you will be able to do

• List what data teams version in Git
• Name secrets and data files that stay out of Git
• See where Git sits on the DCubes roadmap

Why this matters for data work

Pipelines are code. Without Git you cannot answer “what changed Friday?” or review a teammate’s SQL. With Git you can branch, review, and roll forward safely.

What goes in Git

• Python scripts and packages
• SQL files and dbt-style models (later)
• compose.yaml, Dockerfiles, CI config
• README and docs
• Small golden test fixtures (tiny CSVs)

What stays out

• .env and passwords
• cloud keys and tokens
• .venv/ and node_modules/
• large raw data dumps and warehouse extracts
• notebook checkpoint clutter (when possible)

Map of this module

1. Getting Started
2. init / status / diff
3. add / commit / log
4. gitignore
5. clone and navigate 6–8. branches, merge, conflicts 9–11. remotes, GitHub, pull requests
6. undo safely
7. repo hygiene
8. Demo: First Git shift

Practice on your machine

cd ~/dcubes/git-lab
git --version

Next

init, status, diff