Knowing which command to type gets you through a tutorial. Understanding what Git is actually doing gets you through your career. There’s a difference between memorizing commands and seeing your project the way Git does — as a series of saved states. When that mental model clicks, you stop reacting to Git and start reasoning with it. You develop judgment: not just how to use the tool, but how to think about history, state, and collaboration in any system you’ll ever work in. You might forget the commands after a few months, but you will never forget the concepts.
This course builds that understanding from the ground up, across four parts and 64 lectures.
Part 1: How Git Works
You’ll start by understanding Git’s internals — not as a curiosity, but because it changes everything. When you know that a commit is just a snapshot of a folder, that a branch is just a file containing a hash, and that HEAD is nothing more than a pointer, commands like `git reset`, `git rebase`, and detached HEAD state aren’t scary anymore. They’re obvious.
By the end of Part 1, you won’t just know Git commands — you’ll have a mental model that lets you reason through any situation Git can throw at you.
Part 2: Managing History
With a solid foundation in place, you’ll learn to reshape history: moving branch pointers with `git reset`, merging diverged branches, resolving conflicts, stashing work in progress, and rewriting commits with interactive rebase. You’ll understand the real trade-offs between merge and rebase — not just “use rebase for clean history” but *why, when, and at what cost*. You’ll learn how the reflog makes nearly every “lost” commit recoverable.
Part 3: Collaborating with Git
Here’s where Git becomes a team sport. You’ll see how remote repositories work under the hood, why push rejections happen, and what force-pushing actually does to your teammates’ history. A dedicated section on *Developer Velocity* quantifies how your Git decisions multiply — or drain — productivity across an entire team.
Part 4: GitHub
GitHub isn’t just a hosting service — it’s a collaboration platform that solves six concrete problems: access control, branch protection, code review, automation, identity verification, and scaling. You’ll walk through the complete pull request lifecycle (open, review, iterate, merge), build a GitHub Actions CI workflow from scratch, sign commits for the GitHub Verified badge, organize work with Issues and project boards, and manage repositories at the organization level. Then you’ll take the contributor’s side: fork a public repo, open a cross-fork pull request, and watch CI checks gate the merge.
The AI Era
AI tools can reason about Git, and often reason well. The limiting factor isn’t the AI — it’s your ability to evaluate it. You don’t know what you don’t know. You are blind to your own mistakes. Without a deep understanding, you can’t catch a wrong answer, spot a flawed suggestion, or steer toward the right solution. In that interaction, you become the liability, not the asset. A source of confusion, not direction.
