Einstein's inputs were a bunch of scattered theories about electromagnetism and thermodynamics -- particularly, Maxwell's equations for the electromagnetic fiel

Einstein's inputs were a bunch of scattered theories about electromagnetism and thermodynamics -- particularly, Maxwell's equations for the electromagnetic field, and the empirical description of the photoelectric effect, and Planck's description of blackbody radiation. His outputs were deducing models that elegantly explained these phenomena: that a constant speed of light in all reference frames, as unintuitive as that is, would lead to the equations of relativity, and that energy being transmitted only in discrete quanta would lead to the photoelectric effect and blackbody radiation effects that were observed by others. This was largely not a feat of producing theories on data. It was coming up with a simpler explanation for phenomena which had already been known, but for which existing explanations were far too complex. The amazing part is that he did this four times in one year (1905): https://en.wikipedia.org/wiki/Annus_Mirabilis_papers https://en.wikipedia.org/wiki/Annus_Mirabilis_papers