Hi @nickgreenberg. Thanks for your thoughts here. Some great things to consider.
As far as the Spotify thing, BT's thoughts and experience with AI are presented in about the last 15 minutes of that podcast. I've read through a few things about Spotify's "Creator Technology Research Lab" but there isn't a whole to of good info on what exactly their goals are with generative music. I'm not sure if there are any academic papers on the topic and with Spotify being a private company, I'm sure they are keeping data, algorithms, and strategy pretty close to the vest. In general though, it seems like Spotify is a data company masquerading as a jukebox... and they get to keep the money. I can't say that I am a big fan of BT's music, but he definitely knows what he is talking about in regards to music technology and the future.

Regarding drum modules, I have a small subsection at the bottom left dedicated to drums and drum sequencing. I also incorporate an external drum machine here. I've been working more towards synthesizing my own percussion tones with the Patching Panda Punch v3 and low pass gates, and so it makes sense for my workflow to have a drum section integrated within the larger rack. Your recommendation of waveshapers/wavefolders will come in handy here. In fact, I have switched out a few modules since I initially posted the rack image above so I can include a wavfeolder. I haven't decide if I want an Intellijel or Instruo yet. I will definitely try the audio rate switching with the Switchblade. I still have some time to tweak my remaining 104hp before I commit to anything.

And finally, your recommendation for submixers is one that is essential for everyone (and especially noobs) to consider. I have pondered that exact question quite a bit and have ultimately decided to adhere to Jim's insistence over multiple threads that I add a larger matrix mixer (as opposed to smaller submixers). I reasoned that each level of submixing that I incorporate limits the amount of control I have over single sounds at the end of the signal path. If I submix drums, what happens if I decide I want to add reverb to a single snare hit or send just hi-hats to a delay? I could probably use a few more buffered mults, but I have a considerable amount of external passive mults, so signal splitting isn't much of a problem. From a performance perspective I reasoned that a dedicated mixing section worked for me. With the Toppobrillo, Erica Fusion and Drum Mixers, and the Doepfer matrix all in one area of the rack, I can perform with a dub approach that makes sense to me. I can also use the matrix mixer along with the 3xMIA to mix CV in interesting ways, so it seems like a winner. As much as I love the WMD Performance Mixer, I had not initially planned to go this big with the rack, and so I ended up with multiple mixers. I'm actually pretty happy with the feature set and sound quality of the mixing section I've cobbled together. I do have the Moffenzeef Moffenmix (not pictured here) that I can always add back in if I find a need for more mixers.

Thanks again for your thoughts. These are all very useful ideas to ponder.
Take care and enjoy your weekend.