6 ms·

I always wondered about single individuals having an out-sized impact in tech. It is usually system/teams which make things happen. I am skeptical when one pers

I always wondered about single individuals having an out-sized impact in tech. It is usually system/teams which make things happen. I am skeptical when one person is given most of the credit, and ignore all nameless minions who toil long hours to do the real work. Curious to hear others' perspectives on this.

paulcole 6y ago
What you are describing is not somehow unique to “tech.” It’s baked into literally every industry.
fsloth 6y ago
Actually in programming single individuals do make the greatest impact. I'm good at what I do but my work would have far less impact unless it was done in collaboration with the few people in our org whose contributions are astoundingly good. You need lots of peoples, sure. But, at the same time, the outocome of a project actually may depend on the knowhow, skill and diligence of an individual developer. Their output is prodigious, quality sky high, and in this manner they function not only as an individual contributor, but also in a way as a productivity multiplier for everyone they collaborate with, as the standard they set keeps everyone else also striving high. You are correct, in a way - non-collaborating individual contributors are kind of waste and not needed in most cases. But collaborating individual contributors with high standards and high output? You really, really want to work with those.
- amznthrowaway5 6y ago
  In many complex domains like ML, I find most people don't contribute anything and actively make the problem worse. You often don't need the amount of people that large companies often hire, it's just empire building. Software engineers aren't replaceable cogs.
mumpsman 6y ago
I once interviewed with Jim Keller. He was incredibly down to earth and when I pointed out that I wouldn't have ever expected to get an interview with the guy who developed the Athlon 64, he laughed and said that it wasn't he who designed it---it was the huge team of engineers following his guidance. So I got the impression that Jim wasn't looking to take all the credit.
blaser-waffle 6y ago
Standard Pareto 80/20 thing, disproportionate impact and all that. I recall some statistic about how like 4% of fighter pilots accounted for ~40% of the confirmed kills.
eruditegamer420 6y ago
Lmao this is the mantra of all worthless diversity hires. "No one can actually be spectacular, I bet they're all really just ordinary like us."

paulcole 6y ago

What you are describing is not somehow unique to “tech.” It’s baked into literally every industry.

fsloth 6y ago

Actually in programming single individuals do make the greatest impact. I'm good at what I do but my work would have far less impact unless it was done in collaboration with the few people in our org whose contributions are astoundingly good. You need lots of peoples, sure. But, at the same time, the outocome of a project actually may depend on the knowhow, skill and diligence of an individual developer. Their output is prodigious, quality sky high, and in this manner they function not only as an individual contributor, but also in a way as a productivity multiplier for everyone they collaborate with, as the standard they set keeps everyone else also striving high. You are correct, in a way - non-collaborating individual contributors are kind of waste and not needed in most cases. But collaborating individual contributors with high standards and high output? You really, really want to work with those.

amznthrowaway5 6y ago
In many complex domains like ML, I find most people don't contribute anything and actively make the problem worse. You often don't need the amount of people that large companies often hire, it's just empire building. Software engineers aren't replaceable cogs.

mumpsman 6y ago

I once interviewed with Jim Keller. He was incredibly down to earth and when I pointed out that I wouldn't have ever expected to get an interview with the guy who developed the Athlon 64, he laughed and said that it wasn't he who designed it---it was the huge team of engineers following his guidance. So I got the impression that Jim wasn't looking to take all the credit.

blaser-waffle 6y ago

Standard Pareto 80/20 thing, disproportionate impact and all that. I recall some statistic about how like 4% of fighter pilots accounted for ~40% of the confirmed kills.

eruditegamer420 6y ago

Lmao this is the mantra of all worthless diversity hires. "No one can actually be spectacular, I bet they're all really just ordinary like us."