This week I kicked off a new course that's different from anything I've done before. Instead of crafting material from scratch, I went back to a two-day Git course I taught at NASA in 2016 and asked one question of every chapter: how does this hold up in the age of AI-assisted software development?
The course highlights four topics that have changed since 2016: revert and reset, atomic commits, worktrees, and hooks. How do we best use these Git tools while working with AI agents?
Each video ships with a takeaway that you can use in your own projects to adapt Git to modern software development. There is a git-guard script that protects your project from agents running git-reset --hard, a bisect-smoke script that automates git-bisect to find the exact commit where your agent broke your project, and a git-worktree script to make using worktrees simpler. Plus there's a Laravel demo app I use throughout, so you can clone it and break things yourself.
The first video is out now and the remaining four I'll publish next week.
