Putting into words what is off about some of the boxes would probably take quite a big wall of text since it would require explaining perspective again. I do not have the time to do a drawover, so I will try to make it brief. Numer 9-0 and maybe 3.0 look pretty good to me, but almost all other boxes ignore some rules of perspective. This lack of understanding how perspective works and how one of the boxes would have to be drawn in relation to the other, makes it feel like they are not connected properly in many cases. In general i would argue that you have some trouble with making the lines of your boxes converge "correctly" and understanding "3d" space (or rather the illusion of 3d space on a 2d piece of paper). Practicing this with distorted and twisted boxes is making things needlessly more difficult. In other words: You are focusing on landmarks and muscles before having understood perspective enough. Given that you still struggle with perspective you make focussing on landmakrs and anatomy even more difficult. Most of the boxes are also hardly rotated (mostly front facing with some rotation but no bending of 90° or anything) This further makes me feel like you need to work on understanding boxes in perspective more before tackling body shapes. Given that understanding boxes in perspective intuitively is pretty essential, I would recommend to try the 250 Box challenge over on https://drawabox.com/lesson/250boxes . It is pretty boring but it helped me doing boxes intuitively a lot. (read and follow the instructions though, especially the part about checking the convergence of your boxes after every page. If you first draw all the boxes and then check them, you learn a lot less than if you draw five boxes on one piece of paper, check and then implement what you learned on the next page) some words on particular boxes: 1.0 - the lower edge of the top box aligns with the upper back edge of the lower box. Therefor it looks like it is further back. The boxes would have to overlap to be correct. You have improved that in 1.1. 2.0 - The left and right side of the lower box are visible simultaneously, which would only be the case if the box got wider towards the back. In other words, the edges do not converge properly. 12.0 - The edges of the bottom box moving "away" from the viewer are parallel. No real convergence to one point. 15.9 - The top box has all of the edges going "away" from the viewer are also parallel. No real convergence, just some distortion. Whelp… turned into a wall of text anyways. I hope it wasn't too harsh and you can get something out of it. Keep it up.

Getting better, but placement and proportions are still a no-go.

I also find it to be a tricky one - however, it makes sense in my opinion if we consider the central lines (front and back). For the back, we just see half of it, the part until the spine/central line (that we cannot see). Therefore, on your rectangle for the back, you should be able to see half of it. On the front, you can clearly see the central line at the pelvic region, but it then disappears because it is masked by the shoulders. Therefore, the top of your rectangle should cross above the front plane. So I thought it was exaggeration at first, but this is actually accurate. Hope that helps!

i'd really appreciate some feedback, this was quite challenging,,

These are great, you've a great job at capturing the motion of the figure.

These sketches are really well done, I would suggest you try drawing from other reference images that Stan hasn't drawn from before because if you try to deconstruct other images of figures to the robo bean this will help you understand how to identify the blocks on your own.