I worked in Area120 and then at Waymo, in the X building, and saw lots of the other X projects. They built some jaw-dropping stuff. It's tempting to talk about

I worked in Area120 and then at Waymo, in the X building, and saw lots of the other X projects. They built some jaw-dropping stuff. It's tempting to talk about structural problems, incentives, the dynamic between search ads & "moonshots". But I think the main variable that explains the outcomes is quality of management. Running projects like these requires excellent leadership - giving brilliant engineers autonomy to experiment, yet guiding them towards a consistent goal. Knowing when to double down and when to pivot. The stories about John Krafcik, Astro Teller, and Andy Conrad [https://www.statnews.com/2016/03/28/google-life-sciences-exodus/ https://www.statnews.com/2016/03/28/google-life-sciences-exo...], or what happened with Levandowski paint a picture that sufficiently explains the outcomes.