Ok so I just saw Jason’s latest breakdown video:
I just can’t for the life of me figure out how he’s controlling ragdoll, I feel like I’m stuck with some old workflow, every time I use it I gotta adjust poses, then set the timeline back to the start, and re-run the simulation. This looks like he’s literally posing the markers somehow, how does this work? I think I’ve missed some features or something lol
That little lock-button there is what keeps Live Mode from persisting outside of the Manipulator, called “Interactive Mode”.
The rest is him pulling Pin Constraints around, along with a Field towards the end.
I’ll let @Jason fill in the missing bits, this is the first I see of this breakdown myself, you’re fast @MLV! 
Here’s some more behind the scene stuff.
And the final video.
So, basically with Live Mode on and Interactive Mode I can just use pin constraints to move the ragdoll markers around and see in real time what it does? How do you setup poses and keys etc.? Just trying to understand how this new animation workflow actually goes, it’s pretty alien to how most animators work 
I key frame the pins they in essence become my rig, scrubbing backwards will think the sim is still going forwards but once you get the idea of what’s going to happen you can then go out of live mode to adjust.
a short tutorial on this workflow would be great so we can understand it better!
Thank you
Heya @David and welcome to the forums! 
Putting this on my list first thing after the upcoming release of Ragdoll for Blender.
Till then, have a look at our brief documentation around it here:
