5 ms·
by the end of my employment at Google, I was working about 2 hours a day (and getting Meets Expectations at L6, which is OK but not great). While I'd be online
by dekhn 5y ago
by the end of my employment at Google, I was working about 2 hours a day (and getting Meets Expectations at L6, which is OK but not great). While I'd be online and available to respond to chat for 8 hours, I only exerted about 2 hours and most of that was just explaining to executives just how bad our fleet was.
Before that I had worked at least 6 hours a day. And I assumed that to be "right" i should be working at least 8.
From the feedback I got, I wasn't much more productive working more hours. At first I felt guilty and then I realized in my career that working harder isn't going to make any difference in terms of $, career security, or opportunity to work on cooler stuff. So, I just upped my anxiety meds and got back to working 2 hours a day.
- bluefirebrand 5y ago> From the feedback I got, I wasn't much more productive working more hours I've had this feedback too and I feel like it is likely not true. Really what it is is that your managers aren't paying much attention to your output and have sort of written you off for promotions, raises, etc. This was feedback I got a couple of months after I had been working my ass off trying to impress management and then a couple of weeks before getting fired. C'est la vie.
- dekhn 5y agoNo, I'm pretty sure it's true. I had to spend a ton of time explaining what I did and its values to my managers, like writing "ELI5" docs and then suddenly when they "got it" they paid very close attention to my work, and formed a team around me and established scope and got headcount. This is the messaging I got every single quarter: """David, your work is very good. All the people you work with say you're great and it's clear you are adding value, but you need to spend more time finding metrics that demonstrate that value and improving those. Also, if you want to get higher ratings, here is a whole bunch of extra management work and document-writing to do. If you do that, we can make a case for Exceeds and eventually a promotion""" What I concluded was that I was better as an L6 than an L7 (L7 at google has a lot more responsibility, and far fewer job opptys) and I knew the value I was adding to the company without having to spend weeks cooking up metrics dashboards. At google you're not going to be fired, if you're an FTE, with anything less than 6-9 months of feedback and runway to get back in the air (few exceptions)
- francisofascii 5y ago> you need to spend more time finding metrics that demonstrate that value and improving those I have always had trouble with this type of nebulous goal.
- dekhn 5y agoThe challenge for me that I always struggled to explain to my leadership is that in many cases, the time investment to make an accurate metric dashboard greatly exceeds just fixing the problem and knowing anecdotally that it works (I worked with ML Whisperers, people whom I trusted a lot to understand the underlying problems and filter out noise). For example, in my case (sending machines causing invisible data corruption to be replaced) I could have been promoted by doing the following: 1) Finding a metric that correlated with the user pain that I was fixing. In this case, it would be something like "number of jobs that die with a NaN in 1 hour" while running an A/B test (half the jobs in the fleet have some feature enabled) and showign that, with significance, our fix reduces the number of NaNs significantly. (data driven) 2) Demonstrate that the NaN rate corresponds to user productivity (this could be # of papers published, # of models trained per hour, whatever) and that high NaN rates really did have an effect (impact) 3) filter the data carefully, because the vast majority of nans are actually caused by user error, not silent data corruption (this was the actual hard part and nobody has a better solution than "run a determinstic calculation on 8 cores and use majority vote to find the baddie") Run the above for 6 months, show it to all the execs in your division, get a few people from Search, Ads, YouTube or Research/DeepMind to say it increased revenue or decreased costs by 10%. Bingo: promotion, along wiht a full time job maintaining a dashboard with constant requests to add new features, fix code broken by other teams, and making even more presentations to execs on how dysfunctional it is. Or, I could just focus on fixing the machines, hearing anecdotally from the ML Whisperers that it's working again, and go back to surfing hacker news and getting another 100 karma in a day.
- wheelinsupial 5y ago> get a few people from Search, Ads, YouTube or Research/DeepMind to say it increased revenue or decreased costs by 10%. How does that work? Can they genuinely tell that this is the case or do they just give an anecdote or rubber stamp? I’m sure there are dozens/hundreds of other experiments that are going on too. How do you establish the causality at that scale?
- handrous 5y agoI've actually noticed a completely inverse relationship between how hard & how many hours I'm (actually) working and how well I'm perceived by my peers and by management. This is unrelated to acute problems caused by errors on my part, or anything like that. I don't get it, but if I feel like I'm working hard I now take it as a very serious alarm bell. Some of my best feedback has come when I was starting to get nervous because I felt like I was hardly doing anything.
- phkahler 5y agoIf you're working hard at a lot of places you're not getting noticed. If you spend more time talking to people (about work?) They think you're busy. Sometimes the person who is fixing problems is seen as better than the one who is not - it's hard for people to tell which challenges are self inflicted. Sometimes I think the way to get noticed is to spend money. Someone always has to approve your purchases so you get noticed. Sitting in your cube hammering out code, or design, or documentation isnt visible. These things aren't optimal ways to measure effectiveness, but I think they all come into play.
- dekhn 5y agoYour first sentence is actually restating the difference between maker schedules and manager schedules http://www.paulgraham.com/makersschedule.html http://www.paulgraham.com/makersschedule.html
- goodpoint 5y agoIf you work too much you reach a point where you are tired enough that your productivity becomes NEGATIVE. With negative I meant that you might spend 1h to do something that will then take more than 1h to be debugged/reworked/discussed and ultimately removed. Or something that cause breakage that could be avoided. You need to be well rested and clear-minded to foresee a problem and take the right decision, especially around software.
- onlyrealcuzzo 5y ago> At first I felt guilty and then I realized in my career that working harder isn't going to make any difference in terms of $, career security, or opportunity to work on cooler stuff. At least in FAANG, you mostly get paid for what your employer thinks you can do, in theory. Not for what you actually do, in practice. If you're only self-interested - you're much better off spending all your time learning valuable skills and checking boxes to get promoted, and then job hopping - and doing nothing valuable for your employer again and again... You're not going to move up if you spend all your burning through your backlog of work items, and you'll get next-to-no monetary reward or meaningful recognition.