4 ms·

Interesting data, but the focus on the top 5 from each company isn't going to indicate much (too corrupted by geography) and the more informative question is wh

Interesting data, but the focus on the top 5 from each company isn't going to indicate much (too corrupted by geography) and the more informative question is what kinds of positions each school is feeding into. Especially if non-technical positions are included, the locality bias is exactly what we'd expect. Google doesn't demand elite degrees for its lower-level HR functions. My observation is that school pedigree doesn't actually matter that much for getting in the door, but (rather disgustingly) has more of an effect once one is in. It's relatively easy to get a referral that bypasses the resume wall (as long as you're in the geographic area) and then interview performance trumps pedigree. However, pedigree matters a lot after getting the job, especially when it comes to project allocation. If you have pedigree, tech companies know you have other options and will give you decent projects. If you're from SJSU, you might be very good but they'll assume that you don't have other options and you're more likely to get grunt work. Getting the job is the much easier battle in which pedigree doesn't matter much. Getting good projects once in the door is, in large tech companies where it's impossible to prove oneself in a meaningful way until already having favorable project allocation, very strongly influenced by pedigree. People who believe in "meritocracy" will argue that it's possible to succeed on the crap project and gain recognition, but in reality that only gets a person more crap work. This long-lasting and completely inexcusable pedigree effects is one of the dark, ugly secrets that few people will talk about. (People are, morally speaking, OK with the fact that Harvard resumes get more attention than SJSU resumes. The idea that pedigree would still matter among the pool selected for jobs, if better known, would inflame them.)

nostrademons 12y ago
My experience was the opposite - the degree makes them take note of your resume, but once you're in nobody cares (or often knows) where you went to school. FWIW, one of my interviewers went to SJSU. He wrote the JS styleguide, and was a key contributor to Closure and GChat (back when it didn't suck). Another of my interviewers went to UMich/AnnArbor and then toured around the country in a punk rock band; she was the Docs TL before doing a bunch of important work for Search Infrastructure and Chrome. Important work goes to people who seek it out and have a track record of delivering.
- michaelochurch 12y ago
  The problem is that in these large tech companies, the ability to "deliver" is not scarce. Almost everyone is good and the people on bad projects are hideously underemployed, so the rate of "delivering" is 98%. To get a good project, you need a track record of delivering on good projects. Otherwise, you just proved that you're a good grunt. That doesn't buy you anything. Delivering on a bad project just gets you typecast, and that's in the good case where you do well. Then there are the bad projects that are bad (and might not be delivered) because they're impossible or pack the bad kind of challenge (i.e. hideous legacy code, which won't get the same respect as building a new system). The luck of the first project determines everything in the types of companies that we're talking about. And a lot of it is luck: sometimes, as you've noted, people from less prestigious schools land well early on and kill it. However, pedigree is a variable that people (at least, in theory) have some control over. For example, if I wanted to go back and get a PhD, I'm sure I could, and I could probably get that pedigree. (PhD admissions are actually competitive, but I'm certainly smarter than 95+% of those getting into Harvard/Stanford MBA school.) But do I want to deal with an archaic, slow application process and pay 5 years of opportunity cost, given that I'd be 32 at the earliest point that I could start? Probably not, and I shouldn't be penalized for it.
  - nostrademons 12y ago
    Again, my experience was very different from yours, and I believe the difference was due to career choices we made and not strictly random chance. All of the high-priority, high-visibility projects I got because I volunteered for them. My first task at Google was fixing unit tests; my second was writing a signup dashboard for my first project's internal dogfood. It took me about 8 months to end up on a highly-visible project [1], but when I did it got me exec attention, a reputation as someone who was an expert in the department, and a cameo appearance in BusinessWeek. 8 months after you had started at Google, you'd already left in a huff. I know someone who spent 4 years working with the bowel's of Google's cluster management system and transferred successfully to Search, and now to Loon. I know someone who worked for 4 years doing all the grunt work for a project that got canceled; he transferred to Search and now leads a department. I know someone who built his entire career on refactoring "hideous legacy code" - actually, several someones, but the first one I thought of was reliably promoted, every 18 months or so, from SWE 3 to Distinguished, and the other ones are all around Staff or Senior Staff level. Now, whether you're happy or not is a different question, and there are good reasons not to waste 4 years of your life doing something you find boring. But external constraints are not one of them. People judge you based on how well you perform relative to the opportunities in front of you, not what those opportunities are. [1] http://googleblog.blogspot.com/2010/05/spring-metamorphosis-googles-new-look.html http://googleblog.blogspot.com/2010/05/spring-metamorphosis-...

nostrademons 12y ago

My experience was the opposite - the degree makes them take note of your resume, but once you're in nobody cares (or often knows) where you went to school. FWIW, one of my interviewers went to SJSU. He wrote the JS styleguide, and was a key contributor to Closure and GChat (back when it didn't suck). Another of my interviewers went to UMich/AnnArbor and then toured around the country in a punk rock band; she was the Docs TL before doing a bunch of important work for Search Infrastructure and Chrome. Important work goes to people who seek it out and have a track record of delivering.

michaelochurch 12y ago
The problem is that in these large tech companies, the ability to "deliver" is not scarce. Almost everyone is good and the people on bad projects are hideously underemployed, so the rate of "delivering" is 98%. To get a good project, you need a track record of delivering on good projects. Otherwise, you just proved that you're a good grunt. That doesn't buy you anything. Delivering on a bad project just gets you typecast, and that's in the good case where you do well. Then there are the bad projects that are bad (and might not be delivered) because they're impossible or pack the bad kind of challenge (i.e. hideous legacy code, which won't get the same respect as building a new system). The luck of the first project determines everything in the types of companies that we're talking about. And a lot of it is luck: sometimes, as you've noted, people from less prestigious schools land well early on and kill it. However, pedigree is a variable that people (at least, in theory) have some control over. For example, if I wanted to go back and get a PhD, I'm sure I could, and I could probably get that pedigree. (PhD admissions are actually competitive, but I'm certainly smarter than 95+% of those getting into Harvard/Stanford MBA school.) But do I want to deal with an archaic, slow application process and pay 5 years of opportunity cost, given that I'd be 32 at the earliest point that I could start? Probably not, and I shouldn't be penalized for it.
- nostrademons 12y ago
  Again, my experience was very different from yours, and I believe the difference was due to career choices we made and not strictly random chance. All of the high-priority, high-visibility projects I got because I volunteered for them. My first task at Google was fixing unit tests; my second was writing a signup dashboard for my first project's internal dogfood. It took me about 8 months to end up on a highly-visible project [1], but when I did it got me exec attention, a reputation as someone who was an expert in the department, and a cameo appearance in BusinessWeek. 8 months after you had started at Google, you'd already left in a huff. I know someone who spent 4 years working with the bowel's of Google's cluster management system and transferred successfully to Search, and now to Loon. I know someone who worked for 4 years doing all the grunt work for a project that got canceled; he transferred to Search and now leads a department. I know someone who built his entire career on refactoring "hideous legacy code" - actually, several someones, but the first one I thought of was reliably promoted, every 18 months or so, from SWE 3 to Distinguished, and the other ones are all around Staff or Senior Staff level. Now, whether you're happy or not is a different question, and there are good reasons not to waste 4 years of your life doing something you find boring. But external constraints are not one of them. People judge you based on how well you perform relative to the opportunities in front of you, not what those opportunities are. [1] http://googleblog.blogspot.com/2010/05/spring-metamorphosis-googles-new-look.html http://googleblog.blogspot.com/2010/05/spring-metamorphosis-...