I'd like to clear up a couple of what I see as misunderstandings.
Fist, HAZ won't ruin what makes powdered metallurgy special. At its root it is about fine even carbide distribution which really boils down to even alloy distribution. Alloys (the carbide formers) don't really move much, so even if you completely dissolve carbides and totally saturate the matrix, they still reform in basically the same locations.
Powdered metallurgy steels tend to be very clean, which really won't change. And they tend to arrive in a fine grain structure, which is not an inherent property of powdered metal, but is a reflection of the quality of the subsequent processing. This leads me to my second point.
The primary concern about the HAS isn't the hardening and the subsequent need to anneal. It is the extreme grain growth due to the molten temperatures adjacent to the cut. You can anneal the steel to make it workable, but unless you perform grain refinement (which can be a time consuming non-trivial operation) the HAZ from plasma damages the steel some distance from the cut. I would be suspicious of the first 1/8", though it is a function of the cut speed and steel thickness.
In my opinion, the money "saved" doing it yourself with plasma is false economy. You can get a lot of work waterjet for the cost of a CNC plasma machine.