My original background is in mathematics (analysis, topology, Banach spaces) and game theory (imperfect information games). Nowadays, I do AI alignment research (mostly systemic risks, sometimes pondering about “consequentionalist reasoning”).