Happy to share. It’s actually a relatively simple setup with the ER-301.
I chose the samples I wanted to use and loaded them onto a MicroSD card.
On the ER-301 I created a mixer for each sample, and inside each mixer I used either a Sample Player or Manual Grains module, depending on what I wanted to do to that specific sample. I also added fixed delays to some of them, and I think a filter on one, and EQ on another. All of this is inside the Mixer for that sample. All the mixers and samples are in a single unit on the ER-301.
On each Sample Player and Manual Grains instance I assigned CV inputs to slice, grain size, position, trigger, and speed. Those inputs are being triggered and modulated by Ansible Kria, one output of TXo as a noise LFO, NLC 8 Bit Cipher, Modcan Quad LFO, and the SSF URA.
The output from the ER-301 goes into a 4ms DLD, using both channels to create some stereo separation from the mono track.
Out from the DLD into my mixer. An Erbe-Verb is on a mixer send, so I send some of the main stereo track to that and modulate the size, pre-delay, decay, and depth with more outputs from the Modcan and URA.
Also on a send from the mixer is the Cocolase, adding little bits of static and noise bursts along the way.
Once that was all setup I muted all the Mixers in the ER-301, and performed the piece by unmiting tracks, riding the levels, and manipulating the sends, DLD, and Cocolase.
And, not that it’s terrible helpful, here are a few hastily taken photos. 
