We use native and re-designed versions of SOX and OCT transcription factors to reprogram somatic cells into induced pluripotent stem cells and for stem cell engineering. In mammals, what distinguishes paralogous SOX factors is their ability to dimerize with OCT4 on distinctive composite DNA elements to induce stemness (SOX2) or specify the germline (SOX17). The selective partnership of these two SOX factors with OCT4 is primarily directed by a single amino acid conserved across animals. Mutating this amino acid converts SOX17 into an enhanced 'super SOX2', termed eSOX17, which speeds up pluripotency induction. In human cells, eSOX17 enables the direct conversion of somatic cells into totipotent cells. In two-factor cocktails, eSOX17 can transdifferentiate human blood into induced brain stem cells, avoiding the rejuvenation associated with pluripotency induction. Inspired by the close partnership between SOX/OCT and the potential of re-engineered variants for cellular reprogramming, we have begun to study their evolutionary history. Until now, SOX and OCT factors have been believed to be unique to animals. Surprisingly, we identified these factors in certain lineages of unicellular holozoans and found that they can replace SOX2 to induce pluripotency in mice. It is possible that the emergence of the molecular toolkit essential for mammalian pluripotency predates the evolution of multicellularity.