After alot of tinkering, the only thing I could figure out to change on my initial solution was the placement of the critelli. 

Im happy with the method of removing the QS all at once to get to the lead and the assembly of the final molecule. Im not particularly happy with the number of arms required to make and reroute the brick of QS. Feels like a better version of this solution would have 2 less arms and maybe 8 less instructions. I think sub 20 TI is definitely possible. But I couldn't get anything better to work.