I was going to post ^this^. Ted is a badass when it comes to repair. Watching him makes me realize I may have some of the skills but I don't have the temperament.
I do a fair amount of repairs along with a fair amount of building. They are not necessarily one in the same.
I built my first guitar over 15 years ago (and still play it today). It was a kit very much like Pypa just did in the Acoustic forum. There were several reasons for building it but primarily its the fact that I'm an anal engineer and I like to try to understand how things work. There are a whole lot of skills necessary for building an acoustic guitar, doing a kit eliminates some of them, but there was a lot to learn during the process.
Two of the important things you learn when building a guitar is how its structure and geometry works and how to set it up. You buy the tools and learn how to do the work. With luck your guitar is playable and you are happy with it.
Repairs, on the other hand, involves everything that can go wrong with a guitar. Things change with age and playing, things break, things wear out. The most basic part of repairs is adjusting the geometry to be correct and setting it up to be playable. Same tools and skill set as building.
All the rest of repairing is trying to figure out what is wrong and trying to figure out how best to fix it. The head is broken, there is a big crack in the top, the bridge is coming off, binding is loose, it makes funny buzzing sounds..... Erlewine's book is a great reference for how to fix a lot of these individual issues - I have a very dog eared copy on the back of my work bench.
The most important thing to learn about repairs is how to evaluate - I harp on this in my various threads. Take measurements and figure out what they mean. Understand how one affects another. I have a rule that for ANY guitar that appears on my work bench I measure everything before I change anything. I write it down. If its someone else's guitar I talk to them about what I see.
There is another part of repairing that is kind of interesting to think about - when do you not do a repair? I am very careful to consider the age of a guitar, how it was made (meaning what glues), and the impact of my working on it. I'm perfectly capable of doing neck resets, but I don't do them on old Martins because I want that guitar to have a paper trail of who worked on it and what they did. I'll set up a Martin any day of the week but I send the structural stuff to a guy who is really expensive, but worth every penny.
On a similar note, if you work on other people's guitars think about your liability if you screw up. Everyone's guitar is their personal baby, no matter how much you think its a piece of junk.
Speaking of junk, sometimes junk guitars are the way to learn something (your first neck reset for example) but frequently they are frustratingly difficult to work on and often not worth it. I will try to fix anything, but I frequently regret it.
Repairs are the classical way beginning luthiers made money while they tried to build and sell a few guitars. Some find they stay with repair work, it can be a steady income stream.
Whew, that was kind of stream of conscious wasn't it? I guess in summary I would suggest doing both. Build a guitar, kits are fair game, working with a mentor is too. But don't be afraid of doing some fret work or a good setup or replace the pickups in some guitar. Get Erlewine's book, lurk at the repair forum at OLF
Take on a project, get in over your head. Have fun.
ps - I'm always available to lead you astray, er, help.
Freeman Keller, I totally agree with your post. I’m also a “retired” engineer and like to tinker with instruments.........
Find someone who knows how to do it and befriend them.