It boils down to this for me: > in the early days [of rocketry], everyone was a hobbyist The more your company resembles "the early days of rocketry" the more

It boils down to this for me: > in the early days [of rocketry], everyone was a hobbyist The more your company resembles "the early days of rocketry" the more it might make sense to select for hobbyist energy, and substantive side projects are a good signal of that. But: Not all companies resemble or need to resemble the early days of rocketry. They ought to design their own filters based on their needs. Read some Hayek. A monoculture in hiring is undesirable and completely removing bias across all possible dimensions is impossible. For all the talk of "diversity" (read: heterogeneity) in tech, there seems to be a strong cultural bias among engineers toward uniformity and cargo-culting (e.g. the doctrinaire obsession with best practices over first-principles thinking you see in some corners) and little appetite for actual heterogeneity. The inefficiencies emerge when every tiny startup mimics FAANG or every mid-sized company mimics tiny startups, etc.