Asteroid mining is absolutely awesome if you move as much of the process as possible to space - every kilogram you use up there to build something is a kilogram you didn't have to launch upwell.
It'd probably involve some kind of centrifuge design... but I wouldn't rule out just making something screaming hot and then suddenly introducing kinetic energy.
Then there are two main options...
A) Make a crude re-entry shield out of the waste products + some of those other resources you're not keen on harvesting for the primary use of sending to earth, and you put a nice candy core in the middle. Aim the landing trajectory at some shallow ocean.
B) Taking things UP in to orbit is expensive. Taking them back down is probably less so, though our current vehicles aren't based on completing that task so they'd need to changed to handle cargo both ways.
Option C could also be fun... if we ever get space elevators working things could go a lot like in science fiction. (The orbital elevators and start of a planetary ring from Gundam 00 jumps to mind as a reference; 3001 also had the idea much earlier but in less detail)
So maybe you leave the heat shield in orbit, but you've still got get it there.
I think space mining becomes possible when you're no longer cornered by the rocket equation. That is, it's going to require a breakthrough in propulsion systems.
I think near-Earth space mining and manufacturing is what can help us get over that limit in some way - LEO is halfway to anywhere in the Solar System ∆v-wise, so whatever part of your mission you don't have to lift upwell makes for huge savings in fuel (and for many missions, you can use weaker and more efficient engines once you're on a stable orbit).