I've long espoused what jondubois says above and many people have told me what you say here in response. That might be true but now that I'm in a position to see lots of new hires it doesn't seem to make a difference, at least not for CS grads. We hire a bunch of MIT grads and they seem to have roughly the same distribution of success as all of the other new hires. I'll grant you that this is anecdata but it is the one situation where I have a lot of experience where the connection angle does not exist.

It might depend on what you are hiring them for. If it's for research the elites might do better. But if it's for software engineering there might not be a difference.

That's possible, I can only speak to standard SWE type roles, and in those situations the inter-person variance is larger than the inter-institution variance.

Many of the top students at Berkeley, Stanford, MIT don't get standard SWE roles. They go to quant finance and make 300k straight out of undergrad. They go to PhD programs. They start a startup with YC. They do Google APM or something similar. They work as analysts in a VC firm.

I think you learn more from your peer group than classes in college, so it's a good idea got try to get surrounded with the most talented cohort possible.

Exactly. After going to Berkeley, there are 10+ people at top tech companies who think I'm smart and would be more than happy to refer me if I needed a job.

Nice. If I had any sense in my high school years I'd have a Berkeley or Stanford pedigree. Eh, youth is wasted on the young...

Umm... Berkeley also has some of the top CS labs in the world...

I was referring to Berkeley, MIT, and Stanford.