8 ms·

Blind allocation got me too. They told me how allocation worked before I joined, but I was hopeful that they would put me on a project that had some overlap wi

Blind allocation got me too. They told me how allocation worked before I joined, but I was hopeful that they would put me on a project that had some overlap with what I was interested in and what I had worked on for the past 15 years (computer vision, robotics, natural language processing). No. I found out on the last day of orientation I would be doing YouTube ads. I knew there was no way I could do that for 18 months, and I told my manager. He was understanding, but there wasn't really anything he could do. He passed my concerns up the management chain. A couple weeks later my manager gave me their response, which was, literally, "We don't care." That shocked me a little, and I knew at that point I wouldn't be at Google for very long. Maybe that was even the right thing for them to do, since I was proving myself to be the kind of employee who wouldn't work on a project they weren't interested in for a year and a half just because it was good for Google, or good for their potential career at Google. For a short time I led a 20% project with three other engineers that was in my interest area, and after we won an innovation award I hoped that it might make the Google bureaucracy more sympathetic to my preferences, but management didn't care. I left not long after that. (tl;dr: I learned that Google is a bureaucratic megacorp.)

varelse 14y ago
Pretty much my naivete and experience as well (minus the 20% project because the one I wanted to do would have tipped some sacred cows had it succeeded)...
- jjwiseman 14y ago
  Exactly--I didn't quite say it explicitly, but part of the problem was my naiveté about working for a large company after a career of mostly startups. (And my summary of the project may have been overly short. It was a week of full-time work, with a plan of making it an ongoing 20% project.)
  - greenyoda 14y ago
    I don't think this is a large company vs. startup thing. In most companies, large or small, you're hired to fill a particular job opening (that's described as part of the recruitment process) and usually interviewed by the group of people who you'd be working with. Does anyone know of any other companies that do this kind of random assignment the way Google does?
    - chii 14y ago
      I wonder why google would choose to randomize assignments instead of recruiting for a particular skillset for a particular project/role. If they are afraid of leaking information about what projects they are up to based on their job openings, that's actually easy to solve - advertise heaps and heaps of openings for many, varied and niche skill sets. Of course, only some of these adverts are real openings, but no one but outside google will know which ones are real openings. Hence, your competition cannot sniff your job openings to see what you are working on. This simply costs fees in administration and you reject every fake advert application.
      - michaelochurch 14y ago
        Blind allocation is about grunt-work roulette, not secrecy. Google wants to collect smart people, but it's impossible for a company to get to that size and have only smart-people work, so blind allocation is their way of getting PhDs to work on apps for HR-- the kind of work that a lower calibre of engineer would happily do for a Google salary, but Google just refuses to hire at-level for that kind of work when they can hire above-level and roulette someone into it. That's also why it's frowned upon to transfer before the 18-month mark. My issue with it is not that they expect 1.5 years of grunt work. That's fairly typical. Unpleasant, but common. What's worse about it is that people who land on grungy work for their first projects tend to keep getting bad work because they can't transfer to the good stuff, because they're competing against people who had better projects and could actually accomplish things. If the model was, "do this for 18 months, and then the company is open to you", that'd be one thing. The problem is that people suck it up for a year and a half, thinking things will get a lot better, but then when they get to the 1.5-year mark, they don't actually have the degree of optionality and autonomy that they were promised.
michaelochurch 14y ago
20% time can easily destroy your career. Most managers view it either as a waste of time or a flight risk. If the manager has to put someone on the 5% shit-list (which has more to do with his performance than yours) and you're 20%ing, then you're probably screwed. Blind allocation is imbecilic, but I think even the idea that people need to be "allocated" to projects is moronic. It doesn't belong in this fucking century. Closed allocation makes managerial dinosaurs think they control something, but they lose control of the most important thing, which is their ability to retain talent and enthusiasm. If Google wanted to actually deserve its reputation, they'd man up and go for Open Allocation: http://michaelochurch.wordpress.com/2012/09/03/tech-companies-open-allocation-is-your-only-real-option/ http://michaelochurch.wordpress.com/2012/09/03/tech-companie...
- xenen 14y ago
  Wow, I never would have expected Google's perf management to be as bad as what I've read here - Glassdoor reviews all sound so peachy.
  - ori_b 14y ago
    Take what Michael Church says with a grain of salt. I was around to observe his threads on the eng-misc mailing list while he was still at Google. Without his consent, I'd rather not say what went on, but I would advise you to get your information elsewhere.
    - mietek 14y ago
      Michael O'Church seems quite happy to share what went on. What's stopping you? Michael O'Church's arguments are internally consistent, and more persuasive than vague statements impugning his character.
      - trhtrsh 14y ago
        > What's stopping you? Respect for the confidentiality of internal communications. Anyone is free to share their own impressions, but it is inappropriate to pull someone else's privileged documents and wave them in public.
nostrademons 14y ago
I've heard of the 18 month rule, but it wasn't how it worked in my experience. My first project lasted 4 months and ended with a successful launch. My second lasted two months and ended when I was pulled off (actually, I volunteered) to work on a project more suited to my abilities. YMMV. Google is a giant corp, I'm sure there are bad managers somewhere (and I've heard of Nooglers getting stuck on a bad project and then leaving for Facebook), but if you keep your eyes open for opportunities and work to build relationships with your new team, you can usually transfer regardless of what the rules say.

varelse 14y ago

Pretty much my naivete and experience as well (minus the 20% project because the one I wanted to do would have tipped some sacred cows had it succeeded)...

jjwiseman 14y ago
Exactly--I didn't quite say it explicitly, but part of the problem was my naiveté about working for a large company after a career of mostly startups. (And my summary of the project may have been overly short. It was a week of full-time work, with a plan of making it an ongoing 20% project.)
- greenyoda 14y ago
  I don't think this is a large company vs. startup thing. In most companies, large or small, you're hired to fill a particular job opening (that's described as part of the recruitment process) and usually interviewed by the group of people who you'd be working with. Does anyone know of any other companies that do this kind of random assignment the way Google does?
  - chii 14y ago
    I wonder why google would choose to randomize assignments instead of recruiting for a particular skillset for a particular project/role. If they are afraid of leaking information about what projects they are up to based on their job openings, that's actually easy to solve - advertise heaps and heaps of openings for many, varied and niche skill sets. Of course, only some of these adverts are real openings, but no one but outside google will know which ones are real openings. Hence, your competition cannot sniff your job openings to see what you are working on. This simply costs fees in administration and you reject every fake advert application.
    - michaelochurch 14y ago
      Blind allocation is about grunt-work roulette, not secrecy. Google wants to collect smart people, but it's impossible for a company to get to that size and have only smart-people work, so blind allocation is their way of getting PhDs to work on apps for HR-- the kind of work that a lower calibre of engineer would happily do for a Google salary, but Google just refuses to hire at-level for that kind of work when they can hire above-level and roulette someone into it. That's also why it's frowned upon to transfer before the 18-month mark. My issue with it is not that they expect 1.5 years of grunt work. That's fairly typical. Unpleasant, but common. What's worse about it is that people who land on grungy work for their first projects tend to keep getting bad work because they can't transfer to the good stuff, because they're competing against people who had better projects and could actually accomplish things. If the model was, "do this for 18 months, and then the company is open to you", that'd be one thing. The problem is that people suck it up for a year and a half, thinking things will get a lot better, but then when they get to the 1.5-year mark, they don't actually have the degree of optionality and autonomy that they were promised.

michaelochurch 14y ago

20% time can easily destroy your career. Most managers view it either as a waste of time or a flight risk. If the manager has to put someone on the 5% shit-list (which has more to do with his performance than yours) and you're 20%ing, then you're probably screwed. Blind allocation is imbecilic, but I think even the idea that people need to be "allocated" to projects is moronic. It doesn't belong in this fucking century. Closed allocation makes managerial dinosaurs think they control something, but they lose control of the most important thing, which is their ability to retain talent and enthusiasm. If Google wanted to actually deserve its reputation, they'd man up and go for Open Allocation: http://michaelochurch.wordpress.com/2012/09/03/tech-companies-open-allocation-is-your-only-real-option/ http://michaelochurch.wordpress.com/2012/09/03/tech-companie...

xenen 14y ago
Wow, I never would have expected Google's perf management to be as bad as what I've read here - Glassdoor reviews all sound so peachy.
- ori_b 14y ago
  Take what Michael Church says with a grain of salt. I was around to observe his threads on the eng-misc mailing list while he was still at Google. Without his consent, I'd rather not say what went on, but I would advise you to get your information elsewhere.
  - mietek 14y ago
    Michael O'Church seems quite happy to share what went on. What's stopping you? Michael O'Church's arguments are internally consistent, and more persuasive than vague statements impugning his character.
    - trhtrsh 14y ago
      > What's stopping you? Respect for the confidentiality of internal communications. Anyone is free to share their own impressions, but it is inappropriate to pull someone else's privileged documents and wave them in public.

nostrademons 14y ago

I've heard of the 18 month rule, but it wasn't how it worked in my experience. My first project lasted 4 months and ended with a successful launch. My second lasted two months and ended when I was pulled off (actually, I volunteered) to work on a project more suited to my abilities. YMMV. Google is a giant corp, I'm sure there are bad managers somewhere (and I've heard of Nooglers getting stuck on a bad project and then leaving for Facebook), but if you keep your eyes open for opportunities and work to build relationships with your new team, you can usually transfer regardless of what the rules say.