[sterlind]

at my work, I'm a principal and my boss is a partner. he spends a lot of his time doing organizational stuff and coming up with plans to solve problems, but he also still writes PoCs. he'll take the downtime he has between meetings, solve the really hard parts of the problem and hand it to the team for incubation. I do the same thing now - I pitched an idea which requires a fair bit of pure research, and I'm going to get a rough prototype going that vets the idea, then build out my team to make it better and turn it into a robust product.

the trick is to keep the PoCs lean enough that it's easy for your reports to flesh out and change the architecture as they need, but developed enough that it's working code, so people can run it and start working from a grounded product.