8 ms·

What is life like for PhDs in computer science who go into industry?

122 points by gandalfgeek 14y ago

mattdeboard 14y ago
At my previous employer, we hired on a guy who was very close to earning his PhD in CS from a midwestern university known for a prestigious engineering program. I was excited because I thought that at the very least he would be able to learn quickly and start tackling problems relatively quickly. I was not disappointed. He was very assertive about not having his hand held wrt figuring out problems, and was proactive in learning new things. When I left to take another position, I felt comfortable that I was leaving the codebase in good hands.
emmapersky 14y ago
I have a Bachelors in Engineering (BEng) in Computing from Imperial College, not a PhD, but I work in a group at Google where a significant percentage of my peers do, but only know this because from time to time people mention the kind of research they did because either it is relevant to what we are doing, or comes up naturally in discussion. I don't see any difference between what I do, and what others in my group do, we are all in it together, but took different paths to get there. Our group is more research focused (though we are Software Engineers) than many at Google, and this has given me a great opportunity to learn from those with an (even) more academic background. TL;DR - I don't see any differences between PhDs and otherwise at Google, even in 'researchy' teams.
malloc47 14y ago
One very good point this article brings to light is the emphasis academia places on the tenure-track job as the end-all be-all of getting a Ph.D. I can't even begin to count how many times I've heard how getting a professorship is the only worthy job, and how there's no point to a Ph.D. if you don't get such a job. It's nearly to the point of brainwashing (a sweeping generalization, of course, which likely varies greatly from department to department). As a soon-to-be Ph.D. graduate, it's refreshing to know that industry is not only a viable option, but doesn't have the same [superhuman][1] expectations. [1]: http://www4.ncsu.edu/unity/lockers/users/f/felder/public/Papers/Mythpap.html http://www4.ncsu.edu/unity/lockers/users/f/felder/public/Pap...
- javert 14y ago
  I'm 3 years into a PhD at a "second-tier" department (i.e. not MIT or Stanford, but still a pretty good place), and I've never felt any pressure at all to go into academia over industry. So there's one more data point for you. Professors can make a huge "real-world contribution" via their students that go into industry. So they should be proud of those students and feel that the effort of educating them was highly worthwhile.
  - malloc47 14y ago
    It's good to know there are still realistic perspectives in academia. To be fair, the "anti-industry" attitude is probably adviser rather than department dependent, as I have heard a few first-hand data points regarding advisers that align much more with your experience than mine. Given the 1/10 statistic, sooner or later professors will have to satisfy themselves with student's "real-world" contributions, whether they like it or not. I'm just worried how often I see real skill sets (e.g. proper software engineering) being devalued among graduate students because it's not directly relevant to a hypothetical future job as a professor. I feel lucky that I've been able to make extra time to write "real" code from my dissertation work, but it's rather unfortunate that this isn't always the norm.
  - NegativeK 14y ago
    A professor from a local university was giving a computer science related talk; he was quite proud of how many of his PhD students (graduated and ungraduated) had been recruited by Google.
mathattack 14y ago
The incentive system in academia is to produce people who cite their mentors in formal research. That's it. Because universities are non-profits, they go to the indirect incentives. This trumps any altruistic ideals. In their defense, most Phds in computer science are fully funded, so it's not transactional like an undergrad degree. I've worked with several Phds with background in Stats and Computer Science. I found all to have "Good Background" - they knew a lot about their fields. Some were great at working with others, but some not so much. It wasn't a magical degree. Many regretted not getting the 5 years of work experience instead. Relative to the All But Dissertation crowd, they were a little better at Getting Things Done, versus just talking about things. (Very small sample size, so don't read too much into it)
- javert 14y ago
  The incentive system in academia is to produce people who cite their mentors in formal research. Really? My advisor has tenure, and I think he couldn't care less how much his students later go on to cite him. Why should he? He will get lots of citations anyhow (some from former students) because he does a lot of good work, but that's kind of an orthogonal issue.
- sks 14y ago
  Academia has the same incentive system as other fields, i.e., a combination of money/fame/power and all three comes from doing great research. Great research will get you the citations regardless of the number of former students. Also number of citations is a necessary but not a sufficient indicator of great research.
- _delirium 14y ago
  I don't think your analysis of the incentives is accurate at all. If I were to oversimplify similarly, I'd say instead that modern incentives in CS academia, for non-tenured faculty at top research universities, heavily favor money, wherever it comes from (in large part because with budget cuts, professors have no choice). Grants are good, industrial partnerships are good, basically anything that brings in money is good. Citations are secondary, and used mainly as a stepping stone to bring in money. From that perspective, students going to a big company are good, especially a big company like Microsoft, IBM, or Intel that funds research. Students going into academia to write papers that cite the advisor are also good, but not as good as students who go somewhere that has money.
azakai 14y ago
Good points, as a PhD in the industry this all aligns pretty well with my experience as well. The only comment I have is regarding >> Do people usually work alone, or with others? > Again, the answer to this question does not depend on what degree you have. The simple truth is there is no job in the modern tech industry that involves working alone. Agreed that this probably does not depend on what degree you have. But there is a lot of variability in terms of how many people you work with. I worked almost alone for a while, and in large teams at other times. I suppose there is no job where you work 100% alone, with no contact at all with anyone, ever, but I doubt that was the original question.
flatline 14y ago
> there is no job in the modern tech industry that involves working alone. Positions with aspects of management or sales typically involve actually working with people, but this can be hard to achieve for many others. The high tech field can be great for people who want to work in isolation, and fairly lonely for people who do not. I have never worked in a place that effectively implemented things like mentoring programs, for PhDs or anyone else. I am on the East Coast, I suspect that this trend is less prevalent out West. You may form bonds with people through a work environment, but the majority of your daily (and weekly, and monthly) effort is a solitary endeavor. I have heard this complaint from people working on PhD-level industrial research, to coding, to website design. Getting in with a company that recognizes and values real teamwork, assuming you are able to communicate effectively with others, will make nearly any job a much more pleasant experience.
- emmapersky 14y ago
  I think this is really interesting. What "working alone" means is complex. For example, at Hashable, I was effectively working alone, despite there being a wider notion of collaboration. More over, when other people came in to contact with parts of the code I was actively working on, instead of trying to work through those components together (not neccsarily in a pairing way, but just a more general collaboration), the standard was to make you change and not care about the consequences. In a code base without tests this was a nightmare. If / when I am next looking for a job, really understanding how a company collaborates is high up on my agenda of questions.
tensor 14y ago
The only exception I take with this is the part about coding in an interview. My complaints there are not specific to having a PhD though. I understand the need for practical tests like these, but whiteboard coding is not a good way to do it. People dont' typically program on white boards, they do it alone with a compiler, reference manual, and the ability to test. Put them in a room with a terminal and ask them to write a straight forward program in 30 minutes. If you have to practice for the interview, what exactly are you determining in the interview? Their ability to program? Or their ability to practice interview tricks? Immediately discounting a candidate because they are annoyed at your question to whiteboard code also seems rash. Asking someone experienced to write the "print a string in reverse" algorithm will naturally insult them. Especially so if the candidate has a CS PhD and spent time grading students on more complicated programs, or especially if their CV includes links to public repositories containing thousands of lines of code they wrote. On the other hand, asking a candidate to sketch a genuinely complicated algorithm on the board is a reasonable question.
- rm999 14y ago
  >Their ability to program? Or their ability to practice interview tricks? Often, neither. It can be the ability to think clearly and communicate that thinking. I personally avoid whiteboard coding questions when I interview people, but the good ones I've been given involve a good back-and-forth, basically working on the problem together. I usually ask more open-ended questions, like "how would you go about solving this large, potentially ambiguous problem". Both types of questions give the interviewer the ability to gauge personal skills and mental clarity. Whiteboard questions provide some ability to test technical ability in exchange for some ability to test higher-level knowledge. They both work in the same kind of way, though.
- chimmy 14y ago
  also the bit about publications - "it only matters if there are very few or none". Considering the author is a PhD himself, it is some what self-deprecating. And he works at google interviewing PhDs. sigh I believe that quality of publications matters a lot. I've seen many passionate hackers move into industry after Phd and some of them have authored seminal papers. I am positive that most industry positions recognize these contributions. I'm mostly convinced that current form of interviewing process of asking trick programming questions is little random. And this post has many red flags to confirm this. All the best google!
ltcoleman 14y ago
In my experience, the degree has not been a factor at all when it comes to value in industry. I have seen people that didn't have a degree produce more maintainable code in a more productive manner. I also have seen coworkers with bachelors outshine others with masters. I have climbed the corporate ladder at my company and now with my power to influence hires, I rarely take the level of degree in consideration. Strangely, I have seen the attitude problem with new grads that possess a masters or doctorate. I even had one tell my manager that he had a masters and that meant he shouldn't ever have to do support work, and that he was smarter than my manager. He was a fresh 23 year old grad and eventually left for a larger company because he believed they would appreciate his masters. To add salt to the wound, we found that almost every bit of his code either broke our current projects or just flat out wasn't done correctly because he knew better than everybody so there was no reason for him to ask a senior dev or even a BA any questions about what he was doing. All that to say, no matter the shop, no matter the degree, business is about money, and we, as coders, produce value which is why our salaries are high. A coder that is seen as valuable will always do better in there career and most of the time his coworkers will see his value and respect him for that.
bearmf 14y ago
Do you work 9 to 5, or do you take your work home with you? This, again, is completely independent of what degree you have. How you structure your priorities and your work to get it done within sane hours, and while maintaining some sort of “work/life balance” (I hate that term, but that is a whole other story) is entirely up to you. This is indeed independent of degree but highly dependent of company you are working for. There are developers working 10-12 hours a day who cannot structure their work because they are just expected to stay in the office for said time period.
duked 14y ago
I have a PhD and I worked for several by industry research labs. So when the OP mentions " he went “oh well, if you must make me code…” That pretty much made me go “no hire.” He clearly is being either dishonest or taking shortcuts, PhD's want to have a job that is interesting if it's to piss code all day I would have stopped at my Master degree. Pissing code, as much as many would like to think is not exactly challenging if you have a PhD (I'm not saying that having highly optimized, readable code isn't but having something that works is not that hard) so if the whole interview process is centered around how well you can write something in A/B/C/D language yes I understand the interviewee. One funny thing: "Most places will actually pair you up with a mentor who is not your boss" this has to be a joke, I worked for HP-Labs and other big labs never ever had a mentor, usually if you have a issue you are pretty much welcome to discuss it with your colleague, have them review your proposed solution but boy you are on your own when it comes to do implement it. Again the OP has no clue clue of what he is talking and it's sad because his research interests are security related: "You are much more likely to land an interesting job at the coasts. West more than East." I can prove easily that if you have a PhD in a security or somehow related (formal methods) you have more chances to land a job on the east coast: DARPA,IARPA, MITRE, DoD all these agencies are funding 80% of the security research of the country ! Easier if you are a citizen indeed, but I'm a foreigner on H1B and I still work on a DARPA funded project ... One more joke: "Most industrial positions will not care about your publications" seriously this guy is ridiculous or just unable to think outside his google circle, most labs are indeed more than interested in your publications because it means for them potential patent application, more contacts in conferences and the end goal a better reputation and more easier to build a consortium to respond to call for proposals or BAA(http://www.arl.army.mil/www/default.cfm?page=8 http://www.arl.army.mil/www/default.cfm?page=8). But coming from someone who has not many publication I understand, for information here is his publication list http://www.informatik.uni-trier.de/~ley/db/indices/a-tree/h/Haldar:Vivek.html http://www.informatik.uni-trier.de/~ley/db/indices/a-tree/h/... Anyway I'll stop commenting the rest of his post, his post is not all bad but it's based on his personal experience and he does a generalization out of his limited experience.
- jbellis 14y ago
  I've seen approximately zero correlation between degree status and ability to write code. The best programmer I've ever worked with has a PhD... but so have at least two of the terrible ones. I also submit that if programming is "not exactly challenging" then you are working on the wrong products and/or at the wrong level of abstraction. There are legitimate reasons to prefer research, or management, or marketing, or sales over development, but lack of challenge is not one of them.
tokenadult 14y ago
The submitted article contains a key fact, "The problem with this picture is that there are 10 PhD graduates for every tenure-track position. And, while I don’t have figures, the industrial labs don’t hire at a much faster rate either. And that left regular industry jobs as the only viable option for the vast majority of PhDs," which actually applies to a lot of academic major subjects. For the most part, sooner or later a Ph.D. holder may have to consider working for an organization in the for-profit private sector rather than for academia, simply because the private sector is where the majority of the jobs are. I see that the top few comments in this thread as I read the submitted article are mostly about coding tests as a hiring procedure. If the coding test is a reasonably accurate simulation of actual work on the job, it is a very good idea for a company to use a coding test in hiring. From participants in earlier discussions here on Hacker News I have learned about many useful references on the subject of company hiring procedures, which I have gathered here in a FAQ file. The review article by Frank L. Schmidt and John E. Hunter, "The Validity and Utility of Selection Models in Personnel Psychology: Practical and Theoretical Implications of 85 Years of Research Findings," Psychological Bulletin, Vol. 124, No. 2, 262-274 http://mavweb.mnsu.edu/howard/Schmidt%20and%20Hunter%201998%20Validity%20and%20Utility%20Psychological%20Bulletin.pdf http://mavweb.mnsu.edu/howard/Schmidt%20and%20Hunter%201998%... sums up, current to 1998, a meta-analysis of much of the HUGE peer-reviewed professional literature on the industrial and organizational psychology devoted to business hiring procedures. There are many kinds of hiring criteria, such as in-person interviews, telephone interviews, resume reviews for job experience, checks for academic credentials, personality tests, and so on. There is much published study research on how job applicants perform after they are hired in a wide variety of occupations. http://www.siop.org/workplace/employment%20testing/testtypes.aspx http://www.siop.org/workplace/employment%20testing/testtypes... EXECUTIVE SUMMARY: If you are hiring for any kind of job in the United States, prefer a work-sample test as your hiring procedure. If you are hiring in most other parts of the world, use a work-sample test in combination with a general mental ability test. The overall summary of the industrial psychology research in reliable secondary sources is that two kinds of job screening procedures work reasonably well. One is a general mental ability (GMA) test (an IQ-like test, such as the Wonderlic personnel screening test). Another is a work-sample test, where the applicant does an actual task or group of tasks like what the applicant will do on the job if hired. (But the calculated validity of each of the two best kinds of procedures, standing alone is only 0.54 for work sample tests and 0.51 for general mental ability tests.) Each of these kinds of tests has about the same validity in screening applicants for jobs, with the general mental ability test better predicting success for applicants who will be trained into a new job. Neither is perfect (both miss some good performers on the job, and select some bad performers on the job), but both are better than any other single-factor hiring procedure that has been tested in rigorous research, across a wide variety of occupations. So if you are hiring for your company, it's a good idea to think about how to build a work-sample test into all of your hiring processes. Because of a Supreme Court decision in the United States (the decision does not apply in other countries, which have different statutes about employment), it is legally risky to give job applicants general mental ability tests such as a straight-up IQ test (as was commonplace in my parents' generation) as a routine part of hiring procedures. The Griggs v. Duke Power, 401 U.S. 424 (1971) case http://scholar.google.com/scholar_case?case=8655598674229196978&q=Griggs+Duke+Power&hl=en&as_sdt=2,24 http://scholar.google.com/scholar_case?case=8655598674229196... interpreted a federal statute about employment discrimination and held that a general intelligence test used in hiring that could have a "disparate impact" on applicants of some protected classes must "bear a demonstrable relationship to successful performance of the jobs for which it was used." In other words, a company that wants to use a test like the Wonderlic, or like the SAT, or like the current WAIS or Stanford-Binet IQ tests, in a hiring procedure had best conduct a specific validation study of the test related to performance on the job in question. Some companies do the validation study, and use IQ-like tests in hiring. Other companies use IQ-like tests in hiring and hope that no one sues (which is not what I would advise any company). Note that a brain-teaser-type test used in a hiring procedure could be challenged as illegal if it can be shown to have disparate impact on some job applicants. A company defending a brain-teaser test for hiring would have to defend it by showing it is supported by a validation study demonstrating that the test is related to successful performance on the job. Such validation studies can be quite expensive. (Companies outside the United States are regulated by different laws. One other big difference between the United States and other countries is the relative ease with which workers may be fired in the United States, allowing companies to correct hiring mistakes by terminating the employment of the workers they hired mistakenly. The more legal protections a worker has from being fired, the more reluctant companies will be about hiring in the first place.) The social background to the legal environment in the United States is explained in many books about hiring procedures http://books.google.com/books?hl=en&lr=&id=SRv-GZkw6TEC&oi=fnd&pg=PA271&dq=Validity+and+Utility+of+Selection+Models+in+Personnel+Psychology&ots=iCXkgXrlOV&sig=ctblj9SW2Dth7TceaFSNIdVMoEw#v=onepage&q=Validity%20and%20Utility%20of%20Selection%20Models%20in%20Personnel%20Psychology&f=false http://books.google.com/books?hl=en&lr=&id=SRv-GZkw6... http://books.google.com/books?hl=en&lr=&id=SRv-GZkw6TEC&oi=fnd&pg=PA95&dq=Validity+and+Utility+of+Selection+Models+in+Personnel+Psychology&ots=iCXkgXrnMW&sig=LKLi-deKtnP20VYZo9x0jfvqzLI#v=onepage&q=Validity%20and%20Utility%20of%20Selection%20Models%20in%20Personnel%20Psychology&f=false http://books.google.com/books?hl=en&lr=&id=SRv-GZkw6... Some of the social background appears to be changing in the most recent few decades, with the prospect for further changes. http://intl-pss.sagepub.com/content/17/10/913.full http://intl-pss.sagepub.com/content/17/10/913.full http://www.economics.harvard.edu/faculty/fryer/files/Fryer_Racial_Inequality.pdf http://www.economics.harvard.edu/faculty/fryer/files/Fryer_R... http://books.google.com/books?hl=en&lr=&id=frfUB3GWlMYC&oi=fnd&pg=PA9&dq=Validity+and+Utility+of+Selection+Models+in+Personnel+Psychology+%22predictive+validity%22+Duke+Power&ots=5O9Hx_E1vY&sig=g-zERWztBWq3h4guEuv9VVkTh8I#v=onepage&q=Validity%20and%20Utility%20of%20Selection%20Models%20in%20Personnel%20Psychology%20%22predictive%20validity%22%20Duke%20Power&f=false http://books.google.com/books?hl=en&lr=&id=frfUB3GWl... Previous discussion on HN pointed out that the Schmidt & Hunter (1998) article showed that multi-factor procedures work better than single-factor procedures, a summary of that article we can find in the current professional literature, for example "Reasons for being selective when choosing personnel selection procedures" (2010) by Cornelius J. König, Ute-Christine Klehe, Matthias Berchtold, and Martin Kleinmann: "Choosing personnel selection procedures could be so simple: Grab your copy of Schmidt and Hunter (1998) and read their Table 1 (again). This should remind you to use a general mental ability (GMA) test in combination with an integrity test, a structured interview, a work sample test, and/or a conscientiousness measure." http://geb.uni-giessen.de/geb/volltexte/2012/8532/pdf/preprint_j.1468_2389.2010.00485.x.pdf http://geb.uni-giessen.de/geb/volltexte/2012/8532/pdf/prepri... But the 2010 article notes, looking at actual practice of companies around the world, "However, this idea does not seem to capture what is actually happening in organizations, as practitioners worldwide often use procedures with low predictive validity and regularly ignore procedures that are more valid (e.g., Di Milia, 2004; Lievens & De Paepe, 2004; Ryan, McFarland, Baron, & Page, 1999; Scholarios & Lockyer, 1999; Schuler, Hell, Trapmann, Schaar, & Boramir, 2007; Taylor, Keelty, & McDonnell, 2002). For example, the highly valid work sample tests are hardly used in the US, and the potentially rather useless procedure of graphology (Dean, 1992; Neter & Ben-Shakhar, 1989) is applied somewhere between occasionally and often in France (Ryan et al., 1999). In Germany, the use of GMA tests is reported to be low and to be decreasing (i.e., only 30% of the companies surveyed by Schuler et al., 2007, now use them)." Integrity tests have limited validity standing alone, but appear to have significant incremental validity when added to a general mental ability test or work-sample test. http://en.wikipedia.org/wiki/Employment_integrity_testing http://en.wikipedia.org/wiki/Employment_integrity_testing http://apps.opm.gov/ADT/Content.aspx?page=3-06&JScript=1 http://apps.opm.gov/ADT/Content.aspx?page=3-06&JScript=1 http://www.princeton.edu/~ota/disk2/1990/9042/9042.PDF http://www.princeton.edu/~ota/disk2/1990/9042/9042.PDF http://www.hotelschool.cornell.edu/research/chr/pubs/reports/abstract-14602.html http://www.hotelschool.cornell.edu/research/chr/pubs/reports... Bottom line: if you are someone with a Ph.D. degree in an academic subject, congratulations. If you seek a job outside academia with good management that understands research, be prepared to do a work sample test to get the job. Companies that hire on the basis of resume biographical qualifications (what degree you have) do demonstrably worse in hiring than companies that make sure that all job applicants can do the actual work of the job. Wouldn't you rather work somewhere where the company focus is on hiring the capable, rather than on hiring the possessors of school credentials?
justin_vanw 14y ago
I can sum up this article: A PhD is roughly the same as a Masters for getting a job, except that you get paid a bit more if it's a big company with an HR department. You haven't really developed any skills that are useful, and now you are going to be paired up with some 25 year old that has 3 years of experience more than you do at 30. Well, you might be able to do well, or maybe not, some PhD's are great and some are suck, and the same is true for people without degrees. You have to prove you are valuable. This is all going to make you very depressed since you are facing 50 hour weeks doing work you don't really care about, instead of 9 hour weeks doing only what interests you in Academia, but you have to eat so you might as well get on with it. Well, that's my summary :) (I'm a bit biased being a dropout)
- pbiggar 14y ago
  Your "summary" is bullshit and has completely distorted the article. You're not just biased, you're also completely misinformed. > You haven't really developed any skills that are useful, and now you are going to be paired up with some 25 year old that has 3 years of experience more than you do at 30. More likely, you've spent 3-6 years coding, learning to communicate, and work well in a team. > This is all going to make you very depressed since you are facing 50 hour weeks doing work you don't really care about, instead of 9 hour weeks doing only what interests you in Academia. Most recently hired academics on a tenure track work 80 hour weeks. You've got to teach course, mentor research students, write grant proposals, and produce research. If you don't, you're out the door. Imagine I were to suggest that because you're a dropout, all you're good for is saying "do you want fries with that?" That's about the level of what you've done with your comment.
  - bnegreve 14y ago
    I think justin_vanw wanted to show that the article was not really informative. I agree with that, I haven't learnt much by reading it.
    - pbiggar 14y ago
      Oh, if that's the case I overreacted. Sorry justin_vanw!
pbiggar 14y ago
The major distinction I've found from having a PhD is that you come with built-in "social proof". If you're going for a job, you're very likely to get an interview. If you're talking to investors, you're very likely to get past the first triage. After that it's up to you of course, but that's pretty useful.
magnusgraviti 14y ago
Here in Ukraine people write their PhD thesis and work in IT industry because scientist salary is extra low (300$) and average salary in industry is about 1500$+ I found the same situation with PhD status as this article highlights. Degree means nothing here. Really nothing. The only thing which matters are results of your work. PhD should be better in analysis, problem solving etc.
xmpir 14y ago
I think in Europe (at least in Austria) a PhD is a pretty huge factor for a job in industry. In my opinion this is a bad thing, but academic degrees are very important here. This does not especially consider the computer science industry, but it happens there as well. There are some positions that are always taken by academics, for the sake of publicity and reputation. No matter what they can do, the only important thing is the PR-effect of a academic degree.
- barry-cotter 14y ago
  Nowhere on Earth has a higher opinion of academic degrees and higher education than the German speaking world. And if anyplace equals it I don't know where it is.
  - brazzy 14y ago
    Even within the German speaking world, Austria has a reputation of being obsessed with titles.
arido 14y ago
One thing to add: when we interview PhD's we make sure they want to be in industry. There is nothing more off putting when you talk to someone and it becomes immediately clear they much rather would've been in academia but they were not the right person at the right time.
DennisP 14y ago
XKCD's perspective: http://xkcd.com/664/ http://xkcd.com/664/

122 points by gandalfgeek 14y ago

At my previous employer, we hired on a guy who was very close to earning his PhD in CS from a midwestern university known for a prestigious engineering program. I was excited because I thought that at the very least he would be able to learn quickly and start tackling problems relatively quickly. I was not disappointed. He was very assertive about not having his hand held wrt figuring out problems, and was proactive in learning new things. When I left to take another position, I felt comfortable that I was leaving the codebase in good hands.

emmapersky 14y ago

I have a Bachelors in Engineering (BEng) in Computing from Imperial College, not a PhD, but I work in a group at Google where a significant percentage of my peers do, but only know this because from time to time people mention the kind of research they did because either it is relevant to what we are doing, or comes up naturally in discussion. I don't see any difference between what I do, and what others in my group do, we are all in it together, but took different paths to get there. Our group is more research focused (though we are Software Engineers) than many at Google, and this has given me a great opportunity to learn from those with an (even) more academic background. TL;DR - I don't see any differences between PhDs and otherwise at Google, even in 'researchy' teams.

malloc47 14y ago

One very good point this article brings to light is the emphasis academia places on the tenure-track job as the end-all be-all of getting a Ph.D. I can't even begin to count how many times I've heard how getting a professorship is the only worthy job, and how there's no point to a Ph.D. if you don't get such a job. It's nearly to the point of brainwashing (a sweeping generalization, of course, which likely varies greatly from department to department). As a soon-to-be Ph.D. graduate, it's refreshing to know that industry is not only a viable option, but doesn't have the same [superhuman][1] expectations. [1]: http://www4.ncsu.edu/unity/lockers/users/f/felder/public/Papers/Mythpap.html http://www4.ncsu.edu/unity/lockers/users/f/felder/public/Pap...

javert 14y ago
I'm 3 years into a PhD at a "second-tier" department (i.e. not MIT or Stanford, but still a pretty good place), and I've never felt any pressure at all to go into academia over industry. So there's one more data point for you. Professors can make a huge "real-world contribution" via their students that go into industry. So they should be proud of those students and feel that the effort of educating them was highly worthwhile.
- malloc47 14y ago
  It's good to know there are still realistic perspectives in academia. To be fair, the "anti-industry" attitude is probably adviser rather than department dependent, as I have heard a few first-hand data points regarding advisers that align much more with your experience than mine. Given the 1/10 statistic, sooner or later professors will have to satisfy themselves with student's "real-world" contributions, whether they like it or not. I'm just worried how often I see real skill sets (e.g. proper software engineering) being devalued among graduate students because it's not directly relevant to a hypothetical future job as a professor. I feel lucky that I've been able to make extra time to write "real" code from my dissertation work, but it's rather unfortunate that this isn't always the norm.
- NegativeK 14y ago
  A professor from a local university was giving a computer science related talk; he was quite proud of how many of his PhD students (graduated and ungraduated) had been recruited by Google.

mathattack 14y ago

The incentive system in academia is to produce people who cite their mentors in formal research. That's it. Because universities are non-profits, they go to the indirect incentives. This trumps any altruistic ideals. In their defense, most Phds in computer science are fully funded, so it's not transactional like an undergrad degree. I've worked with several Phds with background in Stats and Computer Science. I found all to have "Good Background" - they knew a lot about their fields. Some were great at working with others, but some not so much. It wasn't a magical degree. Many regretted not getting the 5 years of work experience instead. Relative to the All But Dissertation crowd, they were a little better at Getting Things Done, versus just talking about things. (Very small sample size, so don't read too much into it)

javert 14y ago
The incentive system in academia is to produce people who cite their mentors in formal research. Really? My advisor has tenure, and I think he couldn't care less how much his students later go on to cite him. Why should he? He will get lots of citations anyhow (some from former students) because he does a lot of good work, but that's kind of an orthogonal issue.
sks 14y ago
Academia has the same incentive system as other fields, i.e., a combination of money/fame/power and all three comes from doing great research. Great research will get you the citations regardless of the number of former students. Also number of citations is a necessary but not a sufficient indicator of great research.
_delirium 14y ago
I don't think your analysis of the incentives is accurate at all. If I were to oversimplify similarly, I'd say instead that modern incentives in CS academia, for non-tenured faculty at top research universities, heavily favor money, wherever it comes from (in large part because with budget cuts, professors have no choice). Grants are good, industrial partnerships are good, basically anything that brings in money is good. Citations are secondary, and used mainly as a stepping stone to bring in money. From that perspective, students going to a big company are good, especially a big company like Microsoft, IBM, or Intel that funds research. Students going into academia to write papers that cite the advisor are also good, but not as good as students who go somewhere that has money.

azakai 14y ago

Good points, as a PhD in the industry this all aligns pretty well with my experience as well. The only comment I have is regarding >> Do people usually work alone, or with others? > Again, the answer to this question does not depend on what degree you have. The simple truth is there is no job in the modern tech industry that involves working alone. Agreed that this probably does not depend on what degree you have. But there is a lot of variability in terms of how many people you work with. I worked almost alone for a while, and in large teams at other times. I suppose there is no job where you work 100% alone, with no contact at all with anyone, ever, but I doubt that was the original question.

flatline 14y ago

> there is no job in the modern tech industry that involves working alone. Positions with aspects of management or sales typically involve actually working with people, but this can be hard to achieve for many others. The high tech field can be great for people who want to work in isolation, and fairly lonely for people who do not. I have never worked in a place that effectively implemented things like mentoring programs, for PhDs or anyone else. I am on the East Coast, I suspect that this trend is less prevalent out West. You may form bonds with people through a work environment, but the majority of your daily (and weekly, and monthly) effort is a solitary endeavor. I have heard this complaint from people working on PhD-level industrial research, to coding, to website design. Getting in with a company that recognizes and values real teamwork, assuming you are able to communicate effectively with others, will make nearly any job a much more pleasant experience.

emmapersky 14y ago
I think this is really interesting. What "working alone" means is complex. For example, at Hashable, I was effectively working alone, despite there being a wider notion of collaboration. More over, when other people came in to contact with parts of the code I was actively working on, instead of trying to work through those components together (not neccsarily in a pairing way, but just a more general collaboration), the standard was to make you change and not care about the consequences. In a code base without tests this was a nightmare. If / when I am next looking for a job, really understanding how a company collaborates is high up on my agenda of questions.

tensor 14y ago

The only exception I take with this is the part about coding in an interview. My complaints there are not specific to having a PhD though. I understand the need for practical tests like these, but whiteboard coding is not a good way to do it. People dont' typically program on white boards, they do it alone with a compiler, reference manual, and the ability to test. Put them in a room with a terminal and ask them to write a straight forward program in 30 minutes. If you have to practice for the interview, what exactly are you determining in the interview? Their ability to program? Or their ability to practice interview tricks? Immediately discounting a candidate because they are annoyed at your question to whiteboard code also seems rash. Asking someone experienced to write the "print a string in reverse" algorithm will naturally insult them. Especially so if the candidate has a CS PhD and spent time grading students on more complicated programs, or especially if their CV includes links to public repositories containing thousands of lines of code they wrote. On the other hand, asking a candidate to sketch a genuinely complicated algorithm on the board is a reasonable question.

rm999 14y ago
>Their ability to program? Or their ability to practice interview tricks? Often, neither. It can be the ability to think clearly and communicate that thinking. I personally avoid whiteboard coding questions when I interview people, but the good ones I've been given involve a good back-and-forth, basically working on the problem together. I usually ask more open-ended questions, like "how would you go about solving this large, potentially ambiguous problem". Both types of questions give the interviewer the ability to gauge personal skills and mental clarity. Whiteboard questions provide some ability to test technical ability in exchange for some ability to test higher-level knowledge. They both work in the same kind of way, though.
chimmy 14y ago
also the bit about publications - "it only matters if there are very few or none". Considering the author is a PhD himself, it is some what self-deprecating. And he works at google interviewing PhDs. sigh I believe that quality of publications matters a lot. I've seen many passionate hackers move into industry after Phd and some of them have authored seminal papers. I am positive that most industry positions recognize these contributions. I'm mostly convinced that current form of interviewing process of asking trick programming questions is little random. And this post has many red flags to confirm this. All the best google!

ltcoleman 14y ago

In my experience, the degree has not been a factor at all when it comes to value in industry. I have seen people that didn't have a degree produce more maintainable code in a more productive manner. I also have seen coworkers with bachelors outshine others with masters. I have climbed the corporate ladder at my company and now with my power to influence hires, I rarely take the level of degree in consideration. Strangely, I have seen the attitude problem with new grads that possess a masters or doctorate. I even had one tell my manager that he had a masters and that meant he shouldn't ever have to do support work, and that he was smarter than my manager. He was a fresh 23 year old grad and eventually left for a larger company because he believed they would appreciate his masters. To add salt to the wound, we found that almost every bit of his code either broke our current projects or just flat out wasn't done correctly because he knew better than everybody so there was no reason for him to ask a senior dev or even a BA any questions about what he was doing. All that to say, no matter the shop, no matter the degree, business is about money, and we, as coders, produce value which is why our salaries are high. A coder that is seen as valuable will always do better in there career and most of the time his coworkers will see his value and respect him for that.

bearmf 14y ago

Do you work 9 to 5, or do you take your work home with you? This, again, is completely independent of what degree you have. How you structure your priorities and your work to get it done within sane hours, and while maintaining some sort of “work/life balance” (I hate that term, but that is a whole other story) is entirely up to you. This is indeed independent of degree but highly dependent of company you are working for. There are developers working 10-12 hours a day who cannot structure their work because they are just expected to stay in the office for said time period.

duked 14y ago

I have a PhD and I worked for several by industry research labs. So when the OP mentions " he went “oh well, if you must make me code…” That pretty much made me go “no hire.” He clearly is being either dishonest or taking shortcuts, PhD's want to have a job that is interesting if it's to piss code all day I would have stopped at my Master degree. Pissing code, as much as many would like to think is not exactly challenging if you have a PhD (I'm not saying that having highly optimized, readable code isn't but having something that works is not that hard) so if the whole interview process is centered around how well you can write something in A/B/C/D language yes I understand the interviewee. One funny thing: "Most places will actually pair you up with a mentor who is not your boss" this has to be a joke, I worked for HP-Labs and other big labs never ever had a mentor, usually if you have a issue you are pretty much welcome to discuss it with your colleague, have them review your proposed solution but boy you are on your own when it comes to do implement it. Again the OP has no clue clue of what he is talking and it's sad because his research interests are security related: "You are much more likely to land an interesting job at the coasts. West more than East." I can prove easily that if you have a PhD in a security or somehow related (formal methods) you have more chances to land a job on the east coast: DARPA,IARPA, MITRE, DoD all these agencies are funding 80% of the security research of the country ! Easier if you are a citizen indeed, but I'm a foreigner on H1B and I still work on a DARPA funded project ... One more joke: "Most industrial positions will not care about your publications" seriously this guy is ridiculous or just unable to think outside his google circle, most labs are indeed more than interested in your publications because it means for them potential patent application, more contacts in conferences and the end goal a better reputation and more easier to build a consortium to respond to call for proposals or BAA(http://www.arl.army.mil/www/default.cfm?page=8 http://www.arl.army.mil/www/default.cfm?page=8). But coming from someone who has not many publication I understand, for information here is his publication list http://www.informatik.uni-trier.de/~ley/db/indices/a-tree/h/Haldar:Vivek.html http://www.informatik.uni-trier.de/~ley/db/indices/a-tree/h/... Anyway I'll stop commenting the rest of his post, his post is not all bad but it's based on his personal experience and he does a generalization out of his limited experience.

jbellis 14y ago
I've seen approximately zero correlation between degree status and ability to write code. The best programmer I've ever worked with has a PhD... but so have at least two of the terrible ones. I also submit that if programming is "not exactly challenging" then you are working on the wrong products and/or at the wrong level of abstraction. There are legitimate reasons to prefer research, or management, or marketing, or sales over development, but lack of challenge is not one of them.

tokenadult 14y ago

The submitted article contains a key fact, "The problem with this picture is that there are 10 PhD graduates for every tenure-track position. And, while I don’t have figures, the industrial labs don’t hire at a much faster rate either. And that left regular industry jobs as the only viable option for the vast majority of PhDs," which actually applies to a lot of academic major subjects. For the most part, sooner or later a Ph.D. holder may have to consider working for an organization in the for-profit private sector rather than for academia, simply because the private sector is where the majority of the jobs are. I see that the top few comments in this thread as I read the submitted article are mostly about coding tests as a hiring procedure. If the coding test is a reasonably accurate simulation of actual work on the job, it is a very good idea for a company to use a coding test in hiring. From participants in earlier discussions here on Hacker News I have learned about many useful references on the subject of company hiring procedures, which I have gathered here in a FAQ file. The review article by Frank L. Schmidt and John E. Hunter, "The Validity and Utility of Selection Models in Personnel Psychology: Practical and Theoretical Implications of 85 Years of Research Findings," Psychological Bulletin, Vol. 124, No. 2, 262-274 http://mavweb.mnsu.edu/howard/Schmidt%20and%20Hunter%201998%20Validity%20and%20Utility%20Psychological%20Bulletin.pdf http://mavweb.mnsu.edu/howard/Schmidt%20and%20Hunter%201998%... sums up, current to 1998, a meta-analysis of much of the HUGE peer-reviewed professional literature on the industrial and organizational psychology devoted to business hiring procedures. There are many kinds of hiring criteria, such as in-person interviews, telephone interviews, resume reviews for job experience, checks for academic credentials, personality tests, and so on. There is much published study research on how job applicants perform after they are hired in a wide variety of occupations. http://www.siop.org/workplace/employment%20testing/testtypes.aspx http://www.siop.org/workplace/employment%20testing/testtypes... EXECUTIVE SUMMARY: If you are hiring for any kind of job in the United States, prefer a work-sample test as your hiring procedure. If you are hiring in most other parts of the world, use a work-sample test in combination with a general mental ability test. The overall summary of the industrial psychology research in reliable secondary sources is that two kinds of job screening procedures work reasonably well. One is a general mental ability (GMA) test (an IQ-like test, such as the Wonderlic personnel screening test). Another is a work-sample test, where the applicant does an actual task or group of tasks like what the applicant will do on the job if hired. (But the calculated validity of each of the two best kinds of procedures, standing alone is only 0.54 for work sample tests and 0.51 for general mental ability tests.) Each of these kinds of tests has about the same validity in screening applicants for jobs, with the general mental ability test better predicting success for applicants who will be trained into a new job. Neither is perfect (both miss some good performers on the job, and select some bad performers on the job), but both are better than any other single-factor hiring procedure that has been tested in rigorous research, across a wide variety of occupations. So if you are hiring for your company, it's a good idea to think about how to build a work-sample test into all of your hiring processes. Because of a Supreme Court decision in the United States (the decision does not apply in other countries, which have different statutes about employment), it is legally risky to give job applicants general mental ability tests such as a straight-up IQ test (as was commonplace in my parents' generation) as a routine part of hiring procedures. The Griggs v. Duke Power, 401 U.S. 424 (1971) case http://scholar.google.com/scholar_case?case=8655598674229196978&q=Griggs+Duke+Power&hl=en&as_sdt=2,24 http://scholar.google.com/scholar_case?case=8655598674229196... interpreted a federal statute about employment discrimination and held that a general intelligence test used in hiring that could have a "disparate impact" on applicants of some protected classes must "bear a demonstrable relationship to successful performance of the jobs for which it was used." In other words, a company that wants to use a test like the Wonderlic, or like the SAT, or like the current WAIS or Stanford-Binet IQ tests, in a hiring procedure had best conduct a specific validation study of the test related to performance on the job in question. Some companies do the validation study, and use IQ-like tests in hiring. Other companies use IQ-like tests in hiring and hope that no one sues (which is not what I would advise any company). Note that a brain-teaser-type test used in a hiring procedure could be challenged as illegal if it can be shown to have disparate impact on some job applicants. A company defending a brain-teaser test for hiring would have to defend it by showing it is supported by a validation study demonstrating that the test is related to successful performance on the job. Such validation studies can be quite expensive. (Companies outside the United States are regulated by different laws. One other big difference between the United States and other countries is the relative ease with which workers may be fired in the United States, allowing companies to correct hiring mistakes by terminating the employment of the workers they hired mistakenly. The more legal protections a worker has from being fired, the more reluctant companies will be about hiring in the first place.) The social background to the legal environment in the United States is explained in many books about hiring procedures http://books.google.com/books?hl=en&lr=&id=SRv-GZkw6TEC&oi=fnd&pg=PA271&dq=Validity+and+Utility+of+Selection+Models+in+Personnel+Psychology&ots=iCXkgXrlOV&sig=ctblj9SW2Dth7TceaFSNIdVMoEw#v=onepage&q=Validity%20and%20Utility%20of%20Selection%20Models%20in%20Personnel%20Psychology&f=false http://books.google.com/books?hl=en&lr=&id=SRv-GZkw6... http://books.google.com/books?hl=en&lr=&id=SRv-GZkw6TEC&oi=fnd&pg=PA95&dq=Validity+and+Utility+of+Selection+Models+in+Personnel+Psychology&ots=iCXkgXrnMW&sig=LKLi-deKtnP20VYZo9x0jfvqzLI#v=onepage&q=Validity%20and%20Utility%20of%20Selection%20Models%20in%20Personnel%20Psychology&f=false http://books.google.com/books?hl=en&lr=&id=SRv-GZkw6... Some of the social background appears to be changing in the most recent few decades, with the prospect for further changes. http://intl-pss.sagepub.com/content/17/10/913.full http://intl-pss.sagepub.com/content/17/10/913.full http://www.economics.harvard.edu/faculty/fryer/files/Fryer_Racial_Inequality.pdf http://www.economics.harvard.edu/faculty/fryer/files/Fryer_R... http://books.google.com/books?hl=en&lr=&id=frfUB3GWlMYC&oi=fnd&pg=PA9&dq=Validity+and+Utility+of+Selection+Models+in+Personnel+Psychology+%22predictive+validity%22+Duke+Power&ots=5O9Hx_E1vY&sig=g-zERWztBWq3h4guEuv9VVkTh8I#v=onepage&q=Validity%20and%20Utility%20of%20Selection%20Models%20in%20Personnel%20Psychology%20%22predictive%20validity%22%20Duke%20Power&f=false http://books.google.com/books?hl=en&lr=&id=frfUB3GWl... Previous discussion on HN pointed out that the Schmidt & Hunter (1998) article showed that multi-factor procedures work better than single-factor procedures, a summary of that article we can find in the current professional literature, for example "Reasons for being selective when choosing personnel selection procedures" (2010) by Cornelius J. König, Ute-Christine Klehe, Matthias Berchtold, and Martin Kleinmann: "Choosing personnel selection procedures could be so simple: Grab your copy of Schmidt and Hunter (1998) and read their Table 1 (again). This should remind you to use a general mental ability (GMA) test in combination with an integrity test, a structured interview, a work sample test, and/or a conscientiousness measure." http://geb.uni-giessen.de/geb/volltexte/2012/8532/pdf/preprint_j.1468_2389.2010.00485.x.pdf http://geb.uni-giessen.de/geb/volltexte/2012/8532/pdf/prepri... But the 2010 article notes, looking at actual practice of companies around the world, "However, this idea does not seem to capture what is actually happening in organizations, as practitioners worldwide often use procedures with low predictive validity and regularly ignore procedures that are more valid (e.g., Di Milia, 2004; Lievens & De Paepe, 2004; Ryan, McFarland, Baron, & Page, 1999; Scholarios & Lockyer, 1999; Schuler, Hell, Trapmann, Schaar, & Boramir, 2007; Taylor, Keelty, & McDonnell, 2002). For example, the highly valid work sample tests are hardly used in the US, and the potentially rather useless procedure of graphology (Dean, 1992; Neter & Ben-Shakhar, 1989) is applied somewhere between occasionally and often in France (Ryan et al., 1999). In Germany, the use of GMA tests is reported to be low and to be decreasing (i.e., only 30% of the companies surveyed by Schuler et al., 2007, now use them)." Integrity tests have limited validity standing alone, but appear to have significant incremental validity when added to a general mental ability test or work-sample test. http://en.wikipedia.org/wiki/Employment_integrity_testing http://en.wikipedia.org/wiki/Employment_integrity_testing http://apps.opm.gov/ADT/Content.aspx?page=3-06&JScript=1 http://apps.opm.gov/ADT/Content.aspx?page=3-06&JScript=1 http://www.princeton.edu/~ota/disk2/1990/9042/9042.PDF http://www.princeton.edu/~ota/disk2/1990/9042/9042.PDF http://www.hotelschool.cornell.edu/research/chr/pubs/reports/abstract-14602.html http://www.hotelschool.cornell.edu/research/chr/pubs/reports... Bottom line: if you are someone with a Ph.D. degree in an academic subject, congratulations. If you seek a job outside academia with good management that understands research, be prepared to do a work sample test to get the job. Companies that hire on the basis of resume biographical qualifications (what degree you have) do demonstrably worse in hiring than companies that make sure that all job applicants can do the actual work of the job. Wouldn't you rather work somewhere where the company focus is on hiring the capable, rather than on hiring the possessors of school credentials?

justin_vanw 14y ago

I can sum up this article: A PhD is roughly the same as a Masters for getting a job, except that you get paid a bit more if it's a big company with an HR department. You haven't really developed any skills that are useful, and now you are going to be paired up with some 25 year old that has 3 years of experience more than you do at 30. Well, you might be able to do well, or maybe not, some PhD's are great and some are suck, and the same is true for people without degrees. You have to prove you are valuable. This is all going to make you very depressed since you are facing 50 hour weeks doing work you don't really care about, instead of 9 hour weeks doing only what interests you in Academia, but you have to eat so you might as well get on with it. Well, that's my summary :) (I'm a bit biased being a dropout)

pbiggar 14y ago
Your "summary" is bullshit and has completely distorted the article. You're not just biased, you're also completely misinformed. > You haven't really developed any skills that are useful, and now you are going to be paired up with some 25 year old that has 3 years of experience more than you do at 30. More likely, you've spent 3-6 years coding, learning to communicate, and work well in a team. > This is all going to make you very depressed since you are facing 50 hour weeks doing work you don't really care about, instead of 9 hour weeks doing only what interests you in Academia. Most recently hired academics on a tenure track work 80 hour weeks. You've got to teach course, mentor research students, write grant proposals, and produce research. If you don't, you're out the door. Imagine I were to suggest that because you're a dropout, all you're good for is saying "do you want fries with that?" That's about the level of what you've done with your comment.
- bnegreve 14y ago
  I think justin_vanw wanted to show that the article was not really informative. I agree with that, I haven't learnt much by reading it.
  - pbiggar 14y ago
    Oh, if that's the case I overreacted. Sorry justin_vanw!

pbiggar 14y ago

The major distinction I've found from having a PhD is that you come with built-in "social proof". If you're going for a job, you're very likely to get an interview. If you're talking to investors, you're very likely to get past the first triage. After that it's up to you of course, but that's pretty useful.

magnusgraviti 14y ago

Here in Ukraine people write their PhD thesis and work in IT industry because scientist salary is extra low (300$) and average salary in industry is about 1500$+ I found the same situation with PhD status as this article highlights. Degree means nothing here. Really nothing. The only thing which matters are results of your work. PhD should be better in analysis, problem solving etc.

xmpir 14y ago

I think in Europe (at least in Austria) a PhD is a pretty huge factor for a job in industry. In my opinion this is a bad thing, but academic degrees are very important here. This does not especially consider the computer science industry, but it happens there as well. There are some positions that are always taken by academics, for the sake of publicity and reputation. No matter what they can do, the only important thing is the PR-effect of a academic degree.

barry-cotter 14y ago
Nowhere on Earth has a higher opinion of academic degrees and higher education than the German speaking world. And if anyplace equals it I don't know where it is.
- brazzy 14y ago
  Even within the German speaking world, Austria has a reputation of being obsessed with titles.

arido 14y ago

One thing to add: when we interview PhD's we make sure they want to be in industry. There is nothing more off putting when you talk to someone and it becomes immediately clear they much rather would've been in academia but they were not the right person at the right time.

DennisP 14y ago

XKCD's perspective: http://xkcd.com/664/ http://xkcd.com/664/