Guide
GitHub for data practitioners
Learn the Git and GitHub workflow that keeps pipeline changes reviewable, recoverable, and ready to share.
What
Git records changes to the files in a project. GitHub gives that project a shared home, a review workflow, and a history your team can return to. For data work, that means SQL, Python, pipeline definitions, tests, and documentation change together instead of living in a mix of local folders and chat messages.
You will set up Git, connect a project to GitHub, version a small pipeline change, open a pull request, and publish a project you can point to in a portfolio.
How
The guide follows one realistic loop: make a focused change on your computer, validate it in a development environment, inspect the diff, commit it, then send the branch to GitHub for review. The examples use Bruin and dbt, but the Git workflow is the same for any data project.
Before you start
- A terminal and a code editor such as VS Code or Cursor
- A GitHub account (the first step shows you how to create one)
- For steps 3-4, a Bruin or dbt project to practice on - the Local Pipeline Development course sets one up, or use a project of your own
Guide overview
- 1Sign up for GitHub and install GitCreate a GitHub account, open a terminal, install Git, and set the name and email that appear on your commits.
- 2Understand local and remote repositoriesSee how the project on your computer relates to a shared GitHub repository, branches, and commits.
- 3Version a pipeline changeTrack a focused Bruin or dbt change, keep credentials out of Git, and inspect the diff before you commit.
- 4Work with branches and pull requestsUse a branch for a change, push it to GitHub, and open a pull request with the context a reviewer needs.
- 5Build a data portfolio on GitHubPublish a small, runnable project that explains the question, data, models, checks, and commands behind it.
Get help & contribute