Throughout its history, there are examples of people being hired because they were smart and had experience vaguely in the area of electrons, etc. On the other hand, they seem to consistently hire people with good PhD academic pedigrees. PhD academic pedigrees are higher signal than most undergrad pedigrees.
This strong pull recruiting was complemented by, in some situations, enabling people to do a rotation and figure out which group they wanted to work with.
This style of recruiting is very much in line with the “pull” approach. When is push or pull appropriate for innovation orgs?. In terms of ideas going in, Bell Labs required almost none, DARPA requires some, and Venture Research requires extremely well thought-out ideas. Comparing and contrasting Bell labs, DARPA, and Venture Research. It requires a lot more evidence, but this pattern suggests that the longer the ‘gestation’ someone has in an organization before they need to be productive, the more you can get away with getting just “good people.” This conclusion might seem banal, but it’s important for getting organization design and expectations right, and people often screw it up. If you’re going to hire people before they have a role or expertise, you need to have appropriate internal structures in place. Similarly, if you don’t have those structures in place it’s important to have a good external-idea vetting process.