10 ms·

Paul Buchheit interview (2018)

92 points by kurinikku 3y ago

komali2 3y ago
> I only interviewed at one other company and they asked stupid questions like, “name the seven layers of the OSI Networking Stack,” or something that you'd pull out of a textbook, not things that were actually interesting. Feels a bit ironic now. The one time I interviewed at Google it wasn't necessarily stupid but I don't feel like my ability to build product was ever tested. > It became a cycle. These smart people would bring more smart people with them and so on. I think we got the whole systems department at UCSB because Urs was a professor there. It was like pulling on a string of talent and getting all the talent attached to it. I wonder how accessible this "string" is today for most people. I've mostly worked at startups but I was new to the Bay Area and had no network, and it seemed all my coworkers had worked many previous jobs together. Then I left after 6 years and now seems like working at only two companies was perhaps a mistake - the ~50 or so people I worked with are all either at the same companies, or laid off, and aren't really able to "string me along" with them as they move to bigger and brighter things. Maybe in another few years! Or perhaps I'm not good enough to get carried around through startups :P But I'm thinking more along the lines of, the couple of times I was in charge of hiring, I'd open up linkedin or the resume portal, and I'd look at a planet full of hundreds of thousands of probably just fine candidates for the role, five hundred of whom had applied directly, and I'd think damn, I probably should have tried to get into Stanford, when I'm applying blind, I look like just another resume to some overwhelmed hiring manager. I feel like the opportunity to get into the "Next Google" is simply not available to people like me that came out of Houston from a random college, only spent a couple years in the Bay Area, and now live somewhere else random (Taipei). I'm not really complaining, I do fine, and I'm passionate about the things I do work on, but it's hard to deny the allure of getting to work on something that cool. > This is a question which I would be curious for someone else to tell me the answer to! The closest thing I can think of is Bitcoin and crypto. I'm pretty sure they would have captured the attention of 21 year old me. What's different about crypto, though, is that it's full of all these get rich quick scammers, which Linux didn't have. This interview is from 2018, and a few years later there was an absurd boom and crash of cryptocurrencies, where billions of dollars of "value" were created out of thin air in the markets, and then billions more vanished just as quickly. That's very exciting, I agree. But outside of lines going up and down, I'm really curious what someone with Paul's credentials found exciting about cryptocurrencies. I've simply never seen an application that wasn't "database, but worse." All the stuff about liberalizing currency I should think was obviously not going to happen, currencies require the support of governments, if you invent a currency to challenge that government power, governments will simply make your currency illegal, and now your currency is only good for buying drugs and guns, and if your goal is to fight governments, you'll probably find more effective strategies with actual money, or, mutual aid. So I always figured there was just some really exciting things on the engineering side I'm simply not seeing.
- paulsutter 3y ago
  As of that interview, Bitcoin had crashed every year or two since 2011, so the next crash was no surprise at all https://youtu.be/XbZ8zDpX2Mg https://youtu.be/XbZ8zDpX2Mg “Don’t buy Bitcoin, it’s going to crash!!!”
- bookaway 3y ago
  To be fair to Paul Bucheit this was before the "peak crypto grift" period. As a point of reference, the (now) much maligned (on this forum) a16z crypto fund was announced only 3-4 months prior to this interview. A lot of tech angel investors who cut their engineering teeth in web era look for the "what are young nerds like me interested in now" signal that resembles the startup buzz of that early web period. And a lot of talented young people at the time viewed crypto as technically interesting and potentially world changing, with Vitalic as their prophet -- before it was completely taken over by grift. Now you don't necessarily have to believe these technically talented but politically naive and financially ignorant youngsters will successfully usher in a utopia in order to bet on them. You just need to believe that with this many smart kids throwing shit at the wall, it's likely something of lasting value will emerge rather soon. One of my favorite YC videos is Bucheit's Investor Day talk, which covers some of the same ground in this interview, but goes into other topics and has more meat to it [0]. And if you can find it, his post about the passing of his brother and the birth of his child was also rather moving as I recall. Having said that, these days a lot of YC veterans seem to be infected by the political social media circus and don't seem crosscheck their migrating political positions with robust devil's advocate arguments. You have the president of YC posting seemingly more about SF politics than anything startup-related, or at least enough for someone to question their priorities. Even paulg is randomly retweeting videos of SF crime etc. There's a lot more "Don't expect someone to understand something when his salary depends on him not understanding it" level commenting on current events masquerading as objective thoughtful takes from this group for some reason. It's probably very presumptuous of me to say this, and I'm going out on a limb, but I don't think Google employee #23 era Paul Bucheit would go as far as boosting an abjectly cringeworthy cardboard political candidate pandering to the lowest common denominator, who proclaims his #1 most important "TRUTH" commandment to be "God is real". [0] https://www.youtube.com/watch?v=bSbPXrQ9Wf8 https://www.youtube.com/watch?v=bSbPXrQ9Wf8
riku_iki 3y ago
How to be happy after joining Google as employee #23..
- hiddencost 3y ago
  And quitting before it got bad.
kuharich 3y ago
Past comments: https://news.ycombinator.com/item?id=18248525 https://news.ycombinator.com/item?id=18248525
- dang 3y ago
  Thanks! Macroexpanded: Paul Buchheit on Joining Google, How to Become a Great Engineer, and Happiness - https://news.ycombinator.com/item?id=18248525 https://news.ycombinator.com/item?id=18248525 - Oct 2018 (340 comments)
not_your_vase 3y ago
I love the almost excessive humbleness that's just radiating from this interview. But he is just too humble to be excessively humble.
anssip 3y ago
Reading this makes me want switch jobs. I want to again work somewhere where in the morning I wake up and feel excited to work, instead of feeling forced to work on some stuff that does not interest me at all.
- devjab 3y ago
  I’m curious about what is stopping you. I know I’m very privileged to be in a position where there is a lot of doors open, but at least in my part of the world, this seems to be pretty much the common thing for programmers. I personally think it’s what made my adult life possible at all. I have ADHD and I have a very hard time working on things I don’t find interesting to the point where I might not be a functional adult if I couldn’t. So I’m very grateful that I can job hop and work where I want sort of frictionless. I’ve never had a bad experience switching jobs. I’ve been to some interviews where someone better beat me, which sucks, but I’ve also been to several interviews where it was clear that me and them just weren’t the right fit. The latter has never been a bad experience for me, I’ve even made some friends along the way and sometimes I’ve gotten different jobs because the boss I was talking with knew another boss with a better match in their organisation. My best advice is to just get out there and look!
  - ido 3y ago
    To be undiplomatic, most jobs suck - most work for devs is “enterprise” or web development, usually for companies that are either not tech companies or that the tech is not that interesting. Even in the prestige companies like Amazon or Microsoft most devs don’t work on greenfield projects but on maintaining the cash cows.
    - devjab 3y ago
      I guess it depends on what you like. I’m fortunate in the way that I like doing most programming. I love building things, even if they aren’t exciting or if they are using “un-cool” technologies. Heck, I even enjoy maintenance work as long as the end goal is for it to never need human eyes again. What I dislike is typically tied to all the bullshit that comes with programming. Because I’m mentally damaged the way I am, I have to expend an enormous amount of energy to pretend to listen in a daily standup meeting when people I don’t work directly with mill about whatever the code I’m not at all involved with does. Which is how SCRUM “works” in basically any Danish organisation because none of them have teams actually working on the same thing to justify using SCRUM. Or if you bill by the hour, so you can’t help another developer chase down that silly mistake that turns out to be ridiculously obvious because then how do you bill that half an hour in JIRA? Those things are what I personally look to avoid. So yeah, if you’re not that into programming anything, then I guess it’s harder but it’s still possible. The first thing I did at my current job was to work with solar inverters and collect massive amounts of data from our solar plants. Which was certainly a challenge, I haven’t really read a manual the way I needed to read those solar inverter manuals since I did some silly stuff in C with BitMap images during my computer science education. So those jobs are also out there.
neilv 3y ago
> I've never encountered anyone else who combines being an incredible engineer, and being an incredible product person, and an incredible manager, and an incredible designer. > As the PM on Google Maps, he was frustrated that the JavaScript was really slow. [...] So, one weekend he said, “I'm going to rewrite it”. And he rewrote the entire thing, making it 10 times as fast and a third of the size. He completely threw away the code that a team of engineers had been working on for months. Maybe it wasn't as bad as that sounds, but the situation was something like the very ordinary... A team of engineers had been working on a lot of aspects, and optimizing was intentionally put off. Maybe there were also some suboptimal coordination or project management going on, which was hurting the team's effectiveness (which is something a product manager would speak about with the project manager or tech lead). Maybe he did it quietly, as a proof of concept, to confirm his intuition that the team was having problems, and as "data" to quietly show team leader that something is wrong. Or maybe he had a prior OK/encouragement from the team to rewrite the code they'd working on. Alternatively, maybe it was as bad as it sounds -- a PM From Hell, pulling a move that engineers would be WTFing about for years -- but his "incredible" skillset included also being an incredible politician, so team morale wasn't destroyed.
- neilv 3y ago
  Regarding the morale of teams who care deeply about their project, and WTFing about some management move for years later... I was interviewing for a principal-level role at a startup doing something exciting, and one of the engineering leaders told me a story about when they'd been at one of the legendary earlier computer companies. The company had done some WTF thing involving pitting two engineering teams against each other, with awful incentives. (I had a WTF response, and articulated why, and maybe that was a factor in them making me the offer.) There are some morale-destroying engineering/project management moves that people will still be talking about for decades, and which many experienced engineers and managers who hear about it will immediately understand (to some degree) as awful. (As I said in previous message, I suspect that the GMaps move wasn't as bad as it sounded, and hopefully it wasn't one of those WTF stories.)
- devjab 3y ago
  I’m not sure exactly what the PMs role would be in regards to software quality, but from the sounds of it, the engineering team was poorly lead or had unhealthy practices (which likely means it was poorly lead). Which isn’t too uncommon in software engineering mind you, but if that’s not the responsibility of the PM then I don’t really see what is wrong with what Bret Taylor did. If you could improve my code like that, I would welcome it and learn from it. The only people I think should feel “wronged” in that situation is the people who were responsible for an entire team of engineers not being put to better use. Of course unless Bret Taylor worked on that part or the issue, then whatever lead to the production of bad code was never really solved at google maps, but again, unless that’s the PMs area of responsibility then well…
kookamamie 3y ago
A more cynical view would be: be in the right place at the right time. Much of this is hubris, I think - not really something that can attributed to just "doing the right things", nor replicated at scale.
- marban 3y ago
  Underrated comment.

92 points by kurinikku 3y ago

> I only interviewed at one other company and they asked stupid questions like, “name the seven layers of the OSI Networking Stack,” or something that you'd pull out of a textbook, not things that were actually interesting. Feels a bit ironic now. The one time I interviewed at Google it wasn't necessarily stupid but I don't feel like my ability to build product was ever tested. > It became a cycle. These smart people would bring more smart people with them and so on. I think we got the whole systems department at UCSB because Urs was a professor there. It was like pulling on a string of talent and getting all the talent attached to it. I wonder how accessible this "string" is today for most people. I've mostly worked at startups but I was new to the Bay Area and had no network, and it seemed all my coworkers had worked many previous jobs together. Then I left after 6 years and now seems like working at only two companies was perhaps a mistake - the ~50 or so people I worked with are all either at the same companies, or laid off, and aren't really able to "string me along" with them as they move to bigger and brighter things. Maybe in another few years! Or perhaps I'm not good enough to get carried around through startups :P But I'm thinking more along the lines of, the couple of times I was in charge of hiring, I'd open up linkedin or the resume portal, and I'd look at a planet full of hundreds of thousands of probably just fine candidates for the role, five hundred of whom had applied directly, and I'd think damn, I probably should have tried to get into Stanford, when I'm applying blind, I look like just another resume to some overwhelmed hiring manager. I feel like the opportunity to get into the "Next Google" is simply not available to people like me that came out of Houston from a random college, only spent a couple years in the Bay Area, and now live somewhere else random (Taipei). I'm not really complaining, I do fine, and I'm passionate about the things I do work on, but it's hard to deny the allure of getting to work on something that cool. > This is a question which I would be curious for someone else to tell me the answer to! The closest thing I can think of is Bitcoin and crypto. I'm pretty sure they would have captured the attention of 21 year old me. What's different about crypto, though, is that it's full of all these get rich quick scammers, which Linux didn't have. This interview is from 2018, and a few years later there was an absurd boom and crash of cryptocurrencies, where billions of dollars of "value" were created out of thin air in the markets, and then billions more vanished just as quickly. That's very exciting, I agree. But outside of lines going up and down, I'm really curious what someone with Paul's credentials found exciting about cryptocurrencies. I've simply never seen an application that wasn't "database, but worse." All the stuff about liberalizing currency I should think was obviously not going to happen, currencies require the support of governments, if you invent a currency to challenge that government power, governments will simply make your currency illegal, and now your currency is only good for buying drugs and guns, and if your goal is to fight governments, you'll probably find more effective strategies with actual money, or, mutual aid. So I always figured there was just some really exciting things on the engineering side I'm simply not seeing.

paulsutter 3y ago
As of that interview, Bitcoin had crashed every year or two since 2011, so the next crash was no surprise at all https://youtu.be/XbZ8zDpX2Mg https://youtu.be/XbZ8zDpX2Mg “Don’t buy Bitcoin, it’s going to crash!!!”
bookaway 3y ago
To be fair to Paul Bucheit this was before the "peak crypto grift" period. As a point of reference, the (now) much maligned (on this forum) a16z crypto fund was announced only 3-4 months prior to this interview. A lot of tech angel investors who cut their engineering teeth in web era look for the "what are young nerds like me interested in now" signal that resembles the startup buzz of that early web period. And a lot of talented young people at the time viewed crypto as technically interesting and potentially world changing, with Vitalic as their prophet -- before it was completely taken over by grift. Now you don't necessarily have to believe these technically talented but politically naive and financially ignorant youngsters will successfully usher in a utopia in order to bet on them. You just need to believe that with this many smart kids throwing shit at the wall, it's likely something of lasting value will emerge rather soon. One of my favorite YC videos is Bucheit's Investor Day talk, which covers some of the same ground in this interview, but goes into other topics and has more meat to it [0]. And if you can find it, his post about the passing of his brother and the birth of his child was also rather moving as I recall. Having said that, these days a lot of YC veterans seem to be infected by the political social media circus and don't seem crosscheck their migrating political positions with robust devil's advocate arguments. You have the president of YC posting seemingly more about SF politics than anything startup-related, or at least enough for someone to question their priorities. Even paulg is randomly retweeting videos of SF crime etc. There's a lot more "Don't expect someone to understand something when his salary depends on him not understanding it" level commenting on current events masquerading as objective thoughtful takes from this group for some reason. It's probably very presumptuous of me to say this, and I'm going out on a limb, but I don't think Google employee #23 era Paul Bucheit would go as far as boosting an abjectly cringeworthy cardboard political candidate pandering to the lowest common denominator, who proclaims his #1 most important "TRUTH" commandment to be "God is real". [0] https://www.youtube.com/watch?v=bSbPXrQ9Wf8 https://www.youtube.com/watch?v=bSbPXrQ9Wf8

riku_iki 3y ago

How to be happy after joining Google as employee #23..

hiddencost 3y ago
And quitting before it got bad.

kuharich 3y ago

Past comments: https://news.ycombinator.com/item?id=18248525 https://news.ycombinator.com/item?id=18248525

dang 3y ago
Thanks! Macroexpanded: Paul Buchheit on Joining Google, How to Become a Great Engineer, and Happiness - https://news.ycombinator.com/item?id=18248525 https://news.ycombinator.com/item?id=18248525 - Oct 2018 (340 comments)

not_your_vase 3y ago

I love the almost excessive humbleness that's just radiating from this interview. But he is just too humble to be excessively humble.

anssip 3y ago

Reading this makes me want switch jobs. I want to again work somewhere where in the morning I wake up and feel excited to work, instead of feeling forced to work on some stuff that does not interest me at all.

devjab 3y ago
I’m curious about what is stopping you. I know I’m very privileged to be in a position where there is a lot of doors open, but at least in my part of the world, this seems to be pretty much the common thing for programmers. I personally think it’s what made my adult life possible at all. I have ADHD and I have a very hard time working on things I don’t find interesting to the point where I might not be a functional adult if I couldn’t. So I’m very grateful that I can job hop and work where I want sort of frictionless. I’ve never had a bad experience switching jobs. I’ve been to some interviews where someone better beat me, which sucks, but I’ve also been to several interviews where it was clear that me and them just weren’t the right fit. The latter has never been a bad experience for me, I’ve even made some friends along the way and sometimes I’ve gotten different jobs because the boss I was talking with knew another boss with a better match in their organisation. My best advice is to just get out there and look!
- ido 3y ago
  To be undiplomatic, most jobs suck - most work for devs is “enterprise” or web development, usually for companies that are either not tech companies or that the tech is not that interesting. Even in the prestige companies like Amazon or Microsoft most devs don’t work on greenfield projects but on maintaining the cash cows.
  - devjab 3y ago
    I guess it depends on what you like. I’m fortunate in the way that I like doing most programming. I love building things, even if they aren’t exciting or if they are using “un-cool” technologies. Heck, I even enjoy maintenance work as long as the end goal is for it to never need human eyes again. What I dislike is typically tied to all the bullshit that comes with programming. Because I’m mentally damaged the way I am, I have to expend an enormous amount of energy to pretend to listen in a daily standup meeting when people I don’t work directly with mill about whatever the code I’m not at all involved with does. Which is how SCRUM “works” in basically any Danish organisation because none of them have teams actually working on the same thing to justify using SCRUM. Or if you bill by the hour, so you can’t help another developer chase down that silly mistake that turns out to be ridiculously obvious because then how do you bill that half an hour in JIRA? Those things are what I personally look to avoid. So yeah, if you’re not that into programming anything, then I guess it’s harder but it’s still possible. The first thing I did at my current job was to work with solar inverters and collect massive amounts of data from our solar plants. Which was certainly a challenge, I haven’t really read a manual the way I needed to read those solar inverter manuals since I did some silly stuff in C with BitMap images during my computer science education. So those jobs are also out there.

neilv 3y ago

> I've never encountered anyone else who combines being an incredible engineer, and being an incredible product person, and an incredible manager, and an incredible designer. > As the PM on Google Maps, he was frustrated that the JavaScript was really slow. [...] So, one weekend he said, “I'm going to rewrite it”. And he rewrote the entire thing, making it 10 times as fast and a third of the size. He completely threw away the code that a team of engineers had been working on for months. Maybe it wasn't as bad as that sounds, but the situation was something like the very ordinary... A team of engineers had been working on a lot of aspects, and optimizing was intentionally put off. Maybe there were also some suboptimal coordination or project management going on, which was hurting the team's effectiveness (which is something a product manager would speak about with the project manager or tech lead). Maybe he did it quietly, as a proof of concept, to confirm his intuition that the team was having problems, and as "data" to quietly show team leader that something is wrong. Or maybe he had a prior OK/encouragement from the team to rewrite the code they'd working on. Alternatively, maybe it was as bad as it sounds -- a PM From Hell, pulling a move that engineers would be WTFing about for years -- but his "incredible" skillset included also being an incredible politician, so team morale wasn't destroyed.

neilv 3y ago
Regarding the morale of teams who care deeply about their project, and WTFing about some management move for years later... I was interviewing for a principal-level role at a startup doing something exciting, and one of the engineering leaders told me a story about when they'd been at one of the legendary earlier computer companies. The company had done some WTF thing involving pitting two engineering teams against each other, with awful incentives. (I had a WTF response, and articulated why, and maybe that was a factor in them making me the offer.) There are some morale-destroying engineering/project management moves that people will still be talking about for decades, and which many experienced engineers and managers who hear about it will immediately understand (to some degree) as awful. (As I said in previous message, I suspect that the GMaps move wasn't as bad as it sounded, and hopefully it wasn't one of those WTF stories.)
devjab 3y ago
I’m not sure exactly what the PMs role would be in regards to software quality, but from the sounds of it, the engineering team was poorly lead or had unhealthy practices (which likely means it was poorly lead). Which isn’t too uncommon in software engineering mind you, but if that’s not the responsibility of the PM then I don’t really see what is wrong with what Bret Taylor did. If you could improve my code like that, I would welcome it and learn from it. The only people I think should feel “wronged” in that situation is the people who were responsible for an entire team of engineers not being put to better use. Of course unless Bret Taylor worked on that part or the issue, then whatever lead to the production of bad code was never really solved at google maps, but again, unless that’s the PMs area of responsibility then well…

kookamamie 3y ago

A more cynical view would be: be in the right place at the right time. Much of this is hubris, I think - not really something that can attributed to just "doing the right things", nor replicated at scale.

marban 3y ago
Underrated comment.