There's a stringency to the barrier to participation that's missing here. To be in the 95%-ile of heart surgeons, may still mean that you make seemingly obvious

There's a stringency to the barrier to participation that's missing here. To be in the 95%-ile of heart surgeons, may still mean that you make seemingly obvious mistakes, but I kind of doubt it. To be in the 95%-ile of machine-learning engineers ... what does it even mean? In addition to the stringency of the barrier to participation, there's a stringency of the barrier to opportunity for practice. If you take a brilliant heart surgeon out of New York City, and plop them in the middle of Wichita, Kansas, she's going to find a hard upper limit on the number of mitral valve repairs she can do on any given weekend. Similarly, if you take a machine learning engineer who spent the last 10 years at Google and give him a a few thousand data points from a heart health study, his main challenge will be staying busy and his skills will surely degrade without other work.