From the course: Build with AI: Beyond Pair Programming with AI

Prerequisites and how to use the exercise files

From the course: Build with AI: Beyond Pair Programming with AI

Prerequisites and how to use the exercise files

- [Instructor] This is not a project course, and while I strongly encourage you to play along and test out the different tools and techniques covered, it's also not a hands-on coding course. It's more of a, here's how I think you can do things with AI coding tools, and I invite you to try these techniques to see if they work for you, type course. Now, throughout the course, I'll be demoing GitHub Copilot, ChatGPT, Cursor, and Claude. And if you want to follow along in every demo, you need accounts for all these services. That is not a prerequisite, though. And these are also not the only AI coding tools available, so don't feel pressure to get or use any one of them. This course is about a mindset as much as it is about practical application, and you can watch it without using any AI tool. So play along, or sit back and watch, or do a combination. You know what you came here for and how best to reach your goal. I say all this because, on first blush, the GitHub repository with exercise files for this course may appear a bit of a mishmash of unfinished and disconnected pieces, because it is. What I've provided here are samples you can use if you want to, and ignore if you don't, that allow you to play along with me and then play around yourself later, without having to do a huge amount of setup. A fair portion of the demos I go through in this course will use GitHub Copilot in GitHub Codespaces, and that's what this exercise file repo is mostly for. So when you see me in this environment, you can spin up your own Codespace and jump right in to replicate what I'm doing, and I encourage you to do so, because what you'll find is something extraordinary that is central to the purpose of this course. If you do exactly what I do, type in the exact same code or prompts, or click on the same buttons in the same sequence, there's a good chance you'll get a different result from mine. Realizing this and making it an expectation as you work with AI coding tools is essential to your success. Right now, all AI coding tools are implementations of large language models, and these models are probabilistic and non-deterministic. That means the same request can and will produce different results, because most coding problems have multiple different solutions. So to get the most out of this course, go check out the GitHub repository and play along with me when you see me doing demos. Or not, it's entirely up to you.

Contents