My recent post on learning AI for free attracted a lot of interest because demand for learning materials is at an all-time high. In the technology industry, there’s huge demand for continuous education as we try to keep up with new subjects and cross-train to take on different roles. Most developers have an expectation and a desire to keep learning; it’s why we work in software, after all. These days, it’s easier than ever to learn new subjects, with a flood of online resources such as video courses, blogs, and forums to supplement books and online developer documentation. In AI and data science, particularly, there is such a demand for skilled developers that companies like Microsoft and Google are releasing their own training materials. Like many, I’ve taken a few online courses (MOOCs) from the range of different providers. However, just as for classroom learning, one size does not fit all. I find that sitting passively through videos and taking a basic multiple choice quiz at the end of them doesn’t help me learn. Despite the time spent in front of the video, I don’t remember much of the content so, when I need to get coding, it’s hard to get started. The way I learn is through doing — picking a project and cutting code. Many developers learn best from a hands-on approach, where they write code as they learn, break things, and ask questions along the way. This is why I recently wrote a second article about various AI communities to give you ideas for projects and support when you have questions.