What I'm interested in is, what types of jobs/opportunities are you walking away from because you don’t have the GIS (ESRI) tools, analytical and or data management skills? Any assistance would be greatly appreciated.
I might be reading too much (or the wrong things) into your question, but I think it is an interesting starting point for a discussion.
Like you say, the existing software tools are really good at what they do, and there are a lot of people here that are really good at using them. Do potential use cases exist that go beyond what current software tools offer? Or are people interested in using some advanced analytical tools like arcgis offers?
Currently I'm involved in two projects where we are trying to explore machine learning to find specific types of plants in our imagery. This is something the ESRI tools might support (but we don't own them, and I don't know how good they would be at finding our particular plant at the resolutions we are going for.) I sat in on an arcgis webinar the other day and they had a really interesting example of using 'deep learning' to find all the burned out structures in a survey area -- fairly relevant considering all the forest fires out west these day (rake your leaves people!) These are they types of things humans can be pretty good at spotting, but as the jobs scale up in size, it would be really nice to have a computer doing all this dull work of scrolling through possibly 10's or 100's of thousands of images.
In our projects we've bumped into limits or issues with drone deploy and pix4d that have been stumbling blocks for us. As a result we've experimented with writing some of our own stitching/mapping tools from scratch (which I don't recommend other people try, unless you enjoy the same types of pain as I do.)

All our work is released under an open-source (MIT) license in case anyone is interested and wants to experiment on their own data ...
I've just started experimenting with some simple machine learning techniques using python. I can't say I've managed any slam dunks yet (not sure I can even jump up and touch the rim.)

It turns out that what I'm looking for is very subtlety different from it's surroundings in both texture space and color space. Machine learning can do some amazing things, but often not so good at subtle differences. Anyway, I'm trying to learn about machine learning, but apparently I still need to train my brain with a lot more information before I am able to do anything very useful with it.
If there was a place (or people here) that wanted to discuss machine learning and classification tools, I'd be up for it. As I think back, most of my aerial survey work has involved looking for things that are out of place (debris in the ocean, invasive plants, invasive insects ...) Off the shelf software and even really polished dji hardware has tended to almost, but not quite do what I need.
Curt.