I haven't met many engineers who can mentor well, period. My guess is your perception is biased. I know SDEs who prefer conversations remotely. An extreme examp

I haven't met many engineers who can mentor well, period. My guess is your perception is biased. I know SDEs who prefer conversations remotely. An extreme example would be that there's no chance of physical harassment on Zoom. Another would be the "raise a hand" button seems to be more acceptable to use than actually raising a hand in a meeting, with the result that voices are heard that would otherwise not be. And just fundamentally, think of the site you're on. How unimaginative. How big is your data set? Why can't your sample mentor well remotely? Are there bigger data sets or published analyses? What can you do to improve your model? Or is it just that you prefer to work in an office, or you don't trust your employees? Your argument is the "Think of the children^H^H^H interns" argument from the Cathy Merrill's Washingtonian op ed. [1] It doesn't hold water with anyone seriously trying to improve the productivity and happiness of a diverse workforce comprised of people who have different preferences for home or office; in person or zoom. [1] https://www.washingtonpost.com/media/2021/05/07/cathy-merrill-washingtonian-strike/ https://www.washingtonpost.com/media/2021/05/07/cathy-merril...