As a Ph.D. student, one of the questions I hear most is: academia or industry? Every time I meet someone new and they realize that I’m close to graduating, that’s the question I hear. Every time I meet an old friend at a conference, that’s the question I hear. Even when I meet people in my own department who haven’t seen me in a few weeks, that’s often the question I hear. Don’t get me wrong: It’s a good question to ask. I, too, ask it often. I just don’t have a clear answer yet.

Yeah, I know, that’s not ideal. A lot of grad students already know their answer by their 5th year: They’ve done an industry research internship or two and those experiences have either completely sold them on industry or driven them away. Some love it so much that they drop out of the Ph.D. immediately. Others vow to never again leave the hallowed halls of the academe.

My experiences have been less decisive. Despite having worked 5 research internships (2 at Facebook, 2 at Microsoft Research and 1 at Google) and having worked “in academia” [1] at a highly competitive R1 university for nearly 5 years, I still don’t know what I prefer. I like them both [2]. I suspect there are others who feel the same way I do, and I figure my many experiences in industry give me a perspective that few others have before graduating. So, in this series of posts, I’m going to talk about the options and how I feel about them as of this moment. [3]

In this first post, I’m going to lay out the options. In the next post, I hope to speak more about why making the choice is important and my personal experience with facing this decision.

What are the options? [4]

Academia (or, climbing up the ivory tower)

The first option is, of course, staying in academia. Actually, there are several ways one can “stay in academia”. But what people usually mean is getting a tenure-track faculty job at an R1. That is to say, becoming an Assistant Professor at a prestigious university.

If you choose that option, you have at least the following broad responsibilities: fundraising (e.g., applying for grants), Research (e.g., publishing papers), course teaching (e.g., teaching Intro to Computer Science to a bunch of undergrads), individual student mentorship (i.e., advising Ph.D. students), departmental service (e.g., serving on hiring committees) and academic community service (e.g., serving on program committees). The goal is, basically, to excel at all of these things. Then, in 6-9 years, if you can get a handful of other academics to write a letter of recommendation for you claiming that you are the best in the world at what you do, you will probably get tenure.

Once you get tenure, you can supposedly breath a little. But, actually, you probably can’t because you’ll take on even more service responsibilities (e.g., becoming a department head or a Dean or a provost or something).

The other options you have if you want to stay in academia include becoming: a teaching professor, a research professor, a systems scientist, or a post-doc. Often times, being a research professor or a post-doc is stepping-stone to getting a tenure-track position. Teaching professorships are generally not stepping stones but conscious options to focus more, or exclusively, on teaching as opposed to research. I suppose it may be possible to get tenure as a teaching professor, but it’s not as common. Likewise, systems scientist positions are often not stepping stones. Rather, systems scientists are engineers with deeply specialized knowledge who are hired by tenure-track faculty and academic departments to work on specific research projects.

In most disciplines, “staying in academia” is usually considered the ideal option — the one that all students hope for and the one that all advisers nudge their best students towards. This is less true in engineering disciplines, but I think there’s still an implicit vibe that if you are “good” you should be staying in academia, lest you waste your potential. Generally, though, getting an academic job is very difficult. And getting one that matches what you want is even harder.

Consider the fact that there are probably only a handful of universities that excel in your discipline. Then consider the fact that many of those schools will not be hiring in a given year, and, even if they are, they will be hiring only one person. Then consider the fact that the year you will be graduating with your Ph.D., so will hundreds of others who have similar expertise and research interests. So without even considering your specific desires (e.g., location, weather, demographics, quality of students, a good job for your partner), it is already incredibly difficult to get a good academic job.

But for all that toil, there is a pay off: intellectual freedom, schedule flexibility, name recognition, individual impact, the opportunity to cultivate the great minds of the next generation, the best job security any wage earner can get (i.e., tenure), belonging to a community of like-minded great minds, and a sense of pride that you were somehow ever able to get to this point.

Industry (or, jumping off the tower)

The traditional second option is “joining industry.” Joining industry is also not anything specific: it’s a bit of a catch-all phrase that encompasses, basically, taking any job at any company.

Unsurprisingly, then, as with “staying in academia”, there are many kinds of job you can take on if you decide to “join industry”. These jobs vary along several dimensions, but the one most relevant to this discussion is academic-likeness: i.e., how much capital R Research (hereafter just “Research” as opposed to “research”) you get to do. Generally, the understanding is that even if you “join industry”, you will continue doing research in some capacity [5], but you won’t necessarily get to do any Research.

The most academic-like industry position you can get is a Research Scientist position at an industry Research lab (e.g., Microsoft Research, IBM Research or FX Pal). At these labs, you are afforded a lot of freedom. You are often free to pursue any Research problem that matches yours and your manager’s interests. In that way, working at these labs is similar to academia. You just usually get paid more money, probably live in a nicer city, get to outsource a lot of annoying logistics to dedicated administrative staff (e.g., recruiting users, purchasing state-of-the-art equipment), have a nicer office, and don’t have to raise your own funds or be bound by promises you made in your grant proposals. As such, these jobs are typically about as difficult to get as any specific academic job, but there are generally more options available in any given year. This is mostly because industry research labs generally have a rolling call for new hires, so you can, essentially, apply at any time.

The downsides to these jobs are that there is no tenure, you don’t get to teach or work with individual students except on short-term internships, you have a lot less schedule flexibility, and you may, at any time, be forced or highly incentivized to make your research immediately relevant to an extant product. So, while a position may come with a lot of intellectual freedom, that intellectual freedom is not guaranteed nor held sacred in the way it is in academia. It’s just a risk you have to take that some higher-level manager, some day, will decide that “Research” is not worth the investment.

The next step out in the academic-likeness scale is a set of specialized positions that allow you to use your knowledge of Research but make you work very closely with core products: e.g., engineering scientists, user experience researchers or data scientists. These are probably the most common jobs for Ph.Ds who “join industry”. These jobs are typically offered by companies like Facebook, Google and other companies that don’t really have dedicated research labs. They are generally great for having immediate, real-world impact. Typically, you work in or closely with product teams to either create state-of-the-art technologies (e.g., Google’s e-mail autoresponder) or evaluate the effectiveness of and uncover areas of improvement for a product through a combination of qualitative and quantitative analyses. You can also do some Research from time to time, but it’s not as common and it is usually not how you are evaluated.

Then, there are straight-up engineering, managerial or designer positions. The Ph.D., in this case, is just an advanced degree that gets you past a screening filter: It is proof that you are smart and have a lot of domain specific knowledge, much like a Master’s degree. It usually also affords you a higher starting salary and a higher entry level position: e.g., instead of getting hired as a level 3 software engineer, you may get hired as a level 4 or 5 software engineer.

Generally, the pros of industry are that you get paid a lot more money, have better work/life balance, can live in just about any city you would like, and can have more immediate, real-world impact (especially if you work at an end-user facing company like Facebook or Google). The cons are that you have less intellectual freedom, schedule flexibility, individual impact, job security, involvement with the larger academic community and opportunity for close mentorship with students.

Secret options

Actually, there are more than two options. More, in fact, than I’m even going to talk about. Some of these secret options are actually just very specialized versions of industry and academia, but others do not even fit in the dichotomy. So here is a brief walkthrough of some secret options that I thought would be of interest.

Start-ups (or, jumping off the tower, with your stuff, and into a lava pit)

I mean, it’s 2016. I can’t not talk about start-ups. Start-ups are actually just very specialized forms of “joining industry”, but I thought I’d give them their own category because they do not offer the same benefits or downsides as do other “industry” jobs. In fact, arguably, there are more parallels between staying in academia and founding a start-up. As an entrepreneur, you: have a lot of schedule flexibility; experience constant rejection; are always applying for funds; are betting on risky, uncertain ideas; and manage a small group of people who you are deeply familiar with and whose success greatly determines your own. Sound familiar?

Yes, even Ph.Ds can found startups [6]. But, it takes a rare kind of personality to both pursue a Ph.D. *and* be willing to do a start-up. When you think about the canonical start-up founder, you think of people like Mark Zuckerberg: passionate builders who would sooner drop out of college than enroll in graduate school. That’s why it’s so rare to see any Ph.D. students actually go this route.

Still, a lot of STEM Ph.D. students are probably working on algorithms and technologies that are amenable to commercialization. Accordingly, it may be a great option to forego the traditional “academia or industry” route and do your own thing.

Government (or, zip-lining down the tower into a crowd of bureaucrats)

One of the options that breaks the “academia or industry” dichotomy is a government job. There are actually many government agencies that hire Ph.Ds — especially Ph.Ds in STEM fields. These jobs can also vary fairly drastically in academic-likeness. Some jobs are very Research oriented: for example, you could join NASA and actually do rocket science. As I wrote about in a previous post, the NSA hires a number of computer science and math Ph.Ds to advance the state of the art in computer security. Other government jobs have a lot less to do, directly, with Research: for example, you could be part of a committee to review grant applications at the NSF, or a policy maker at the FTC. These jobs all require knowledge of and experience with conducting Research but do not necessarily require you to do any original Research yourself.

Government jobs usually have great benefits: pension plans, work/life balance, health insurance options, vacations days. In addition, you have may have a lot of influence in far reaching policy decisions. Plus, you might take some pride in knowing that you are a public servant. But, government jobs also mean that you put yourself right in the thick of a slow-moving bureaucracy. How does this affect you? Maybe not a lot. But, you will probably be forced to use highly out of date software (as all government software needs to be heavily audited), you will probably need to fill out a lot of paperwork to do things like conference travel, and I imagine that it would be difficult to publish. Also, government jobs usually pay a lot less than industry jobs because of the government salary cap.

Academia abroad (or, gliding into a tower far, far away)

Another option worth mentioning is “staying in academia”, just not in the US. While US universities are widely considered to be some of the best in the world, there are many universities in Europe and Asia that rival US universities in Research productivity. Often, academic systems vary tremendously across countries. So, if you don’t like the US system where you must self-fund your salary and your lab activities through grants, it’s possible to still find a job in academia elsewhere. I won’t speak too much about this option, however, because I don’t know enough about academic systems in other countries to speak confidently.

There are, of course, many other secret options: for example, consulting, education administration, or joining the public school system. But, in the interest of not completely boring you, I think I’ll stop here.

Parting Words

As a graduating Ph.D. student, choosing a job is probably one of the biggest decisions you’ll have to make. Typically, the question you’re answering is “academia or industry?” but, as I hope I’ve convinced you through this post, there is actually a lot of variety to both of those options and there are options outside of that dichotomy as well. In the next post, I hope to speak more about the implications, on your career, of selecting any one of these options. As well, I hope to share some of my own thoughts on the decision.

One final note: If you liked this post and would like to show your support, here are two things you can do:
– Follow me on Twitter @scyrusk; and,
– Consider signing up for my mailing list below. I won’t spam, you can unsubscribe at any time, and you’ll get some content that I don’t post on the blog.

Email Address:

If you do either or both of those things, you’d make me happy. Thanks!

Footnotes

[1] This is in quotes because being a professor is, actually, very different from being a graduate student from what I gather. Being a graduate student is sort of like an apprenticeship in the sense that you are learning how to do research from someone who already knows who to do it (in theory). But, it’s not at all like an apprenticeship in that the person who is teaching you is rarely doing the things you are doing. They (the professors) are typically doing things you have either rarely or never done: writing grants, serving on program committees, teaching full classes with no one’s help, giving talks about “research programs” that have nothing to do with the specific execution of any one research project. In that sense, getting a Ph.D. doesn’t really prepare you for being a professor — it just teaches you how to do research, which every professor should know how to do.

[2] This really isn’t a cop-out. It’s true. There are some things that are awesome about academic life (mostly: flexibility and freedom), and there are other things that are awesome about industry life (e.g., work/life balance, generally less pressure).

[3] Within reason, of course. I’m going to try and not sabotage my professional options through this blog post!

[4] Okay, just as a disclaimer: I can guarantee that people will not agree with what I have to say from this point on. I’m sure, for example, that my description of the responsibilities of a professor is off the mark to some degree or another. Please remember: my information comes strictly from observation and conversation, as I have never actually held a full-time post-doctoral position.

[5] Research is, after all, what a Ph.D. education is supposed to train you to do.

[6] By the way, my choice of the verb “found” is very intentional. It is also possible to “join” a start-up, but unless you are a very early employee, just joining a start-up is probably not that different than taking a non-research position at any larger company.