Comment by mayanksinghal | Hacker News Reader

Comment by mayanksinghal | Hacker News Reader

Parent Full thread

Pardon me if I am wrong, because I believe I am more clueless than most, but the reasons VCs invest in their MBA buddies may not be because that is an easy (as in lazy) decision but because it is a safe decision. Traditional coaching does ensure that the candidate has at least his/her basics clear. Of course, the decision has to take into account the necessary skills as well, there is no denying that, but Ivy League should be guaranteeing some minimum competency.

> .. Silicon Valley is how much nepotism and old-boy-network

If I am ever in a position to hire (and affect through it entirely and only my own economic well being) I wouldn't mind having people I enjoy company of and can relate to. Genuinely awesome people are a plus point, but even in that category if I am able to find people who share the same background as mine, is it really that bad a decision?

> It was an absolute disaster but my own fault for taking this as proxy for being good on a startup

But there are bad apples in all baskets. I am sure there would be some evidence of great hackers without Ivy League credentials being poor matches for some startups. And I am not arguing that credential replace the need for a test of actual skills, interest and commitment; I am just arguing that the probability of the candidate possessing some minimum value for all three points is fairly high.

Easy is the same as appearing safe - potato potato - your argument for it BEING safe however is nonsense... VC returns don't come from minimum competency. They come from a disproportionate number of home runs. Funding for familiarity with the school is absolutely NOT making the best investment with the money for which you are responsible, because it's not a decision criteria for putting risk capital to work - it's a risk avoidance strategy. (As you seem to instinctively see by your use of the word safe.)

The question isn't whether "if all else is equal should I hire someone like me" it's whether you should prioritize a certain credential over other factors, as happens in a lot of hiring. But to come back to your "is it really that bad a decision?" I think most evidence points to YES, increased diversity of background and opinion benefits decision making groups - in this case companies companies - more than homogeneity, with its tendency towards systematic error.

To challenge your final point given the fact that a lot of the top schools have large numbers of legacy hires, and getting into them often involves far more factors than mere merit you're hoping that the delta in better schooling techniques outweighs recruitment drawbacks. Maybe, but I think it's far from proven.

Nobody is saying DON'T hire people from top schools, with great reputations, and track records of turning out quality candidates. Of course not. What's being said is that optimizing your hiring process to a handful of schools, given how many others are doing the same, makes it unlikely you're getting the very best candidates on the market - I don't see how this is anything other than statistically self-evident.

> If I am ever in a position to hire (and affect through it entirely and only my own economic well being) I wouldn't mind having people I enjoy company of and can relate to. Genuinely awesome people are a plus point, but even in that category if I am able to find people who share the same background as mine, is it really that bad a decision?

Yes, it's the opposite of a meritocracy, which i'm not sure if there's a specific word for. Very close to cronyism.

> It's the opposite of a meritocracy ... You seem to ignore the part where I mention that the two candidates are of equivalent if not equal skill set. If pure technical expertise was all that was expected from people, we wouldn't be singing songs about people-skills, entrepreneurial spirit and cultural _fitness_ like we do. Most of these values decide only how well the person is going to gel with the team.

No two candidates are ever equivalent.

Reply on news.ycombinator.com