Interesting, my path is a bit like the opposite. I tend to avoid categories of tools that abstract too much the problem in a "magic" way for the same reason: you can't easily understand what's going on behind the scene, and you have to dive in each time you enter a corner case. If you can't control what it is doing on each step, then you can't be sure it will be doing what you expect and this can become a mental effort that outweight the tool benefits.