Don't get me wrong, CMU has a great computer science faculty, but looking at random CMU alumni doesn't tell you much because a random CMU grad isn't that different from a random Georgetown grad.
I think this article chose CMU because for Computer Science and technical startups, it's well understood in the valley that CMU is a top 4 school (and for many, a top 2 school) for producing software engineers, which tend to underpin a lot of tech startups.
Totally agree if you move outside of software companies that this makes sense. Anecdotally, I'll say CMU tends to produce individuals who are technically overweight but perhaps lacking in business sense. This is both because of school culture and the broader Pittsburgh community.
Also, as with CMU, there are definitely more people who are interested in the topic, and a subset of them really are quite talented, but the talent density just isn't as high as at HYPS/MIT/Caltech, and as you noted, the talented people aren't as well-rounded (because if they were, they would've gotten into HYPS/MIT/Caltech).
My top 4 would be mit, Berkley, Stanford, and caltech.
I've probably never looked at the edu section of any CV.
But agreed that the point of this dataset is not accuracy or explanatory power... it's to drive some clicks to a $4.99 Stripe flow and to "an AI powered data-driven market intelligence platform for private growth stage investors". HN loves eating up content-marking.
I'd probably also argue that the headline is at least suggestive that the data is adjusted for school size and I don't believe it is. (Again, not clear if it should be or not but should be clear that this is about aggregates at the institution level.)
University of California, Berkeley
University of California, Berkeley, Haas School of Business
University of California, Berkeley - Walter A. Haas School of Business
UC BerkeleyWhile the top 5% of Stanford grads might be "better" due to how selective they are. But I'd bet the median CMU CS major knows more, programs better etc. Simply because Stanford has a high percent of legacy, athlete, and moneyed international students that buy in. Whereas anecdotally, while CMU students are less well-rounded, those students are serious grinders.
If anything, this is proof that prestige, not talent, and who you know matters most. I've heard of several start ups with mostly Berkley/UC grads and one Stanford grad, and they always introduce him first for pitches.
It’s only obvious because they are given the chance to demonstrate their capability.
I’m arguing that the inherent baseline quality of student does not differ much. If given the chance CMU/Berkeley/etc alums can also succeed. But they’re locked out of many chances due to school prestige.
Yes I do agree that I could have done a better job at removing duplicates like "UC Berkeley" and "University of California, Berkeley" from the figure but the key takeaway wouldn't have changed.