[hkmurakami]

I've worked in Japan in a mega Corp R&D division.

There are candidate pools that are "potential hires" with no assignment pre offer, and there are "specialist hires" where you know the team you'll be assigned to. Only accept the latter.

The job hunt process isn't nearly as bad for technical hires as it is for humanities grads.

I completely agree that grad level work should be done in English. I know a few brilliant researchers at Stanford who will likely head back to Japan to take academic posts because their English (and mainly politics of academia) skills just weren't good enough, and their taking the "wiser route out".

I don't know your field, but I hear good things about places like Hitachi Chuou Research.