It's funny, because my experience is almost precisely the inverse. I found that "The Art of Electronics" includes a lot of very suboptimal circuits that easily

It's funny, because my experience is almost precisely the inverse. I found that "The Art of Electronics" includes a lot of very suboptimal circuits that easily mislead people. There are a lot of "restricted domain" circuits in AoE and very little guidance saying "If you're using this, you probably screwed up." By contrast, I learned and built stuff from the Mims circuits all through high-school. Things worked fine for me. Now, I didn't try to do RF circuits--I was mostly doing lower speed analog with some digital stuff. I don't remember any of it being flaky. In fact, I remember using one of my Mims handbooks to help debug one of my AoE circuits. I hit "The Art of Electronics" in college via my Physics department starting freshman year (I was a mainline EE taking a bunch of Physics classes--don't ask). Putting together experiments using circuits from "The Art of Electronics" was generally a lot of frustration. Of course, like you, once I hit the real "Electronics 101" (a Junior year class), those problems all went away. Suddenly I knew which ones would work for which domains and my frustration mostly went away.