How to start an academic career after 5-6 years in industry, without an undergraduate degree?

Question

As an undergraduate, I wasted a ridiculous amount of time due to a combination of being distracted by part-time work, lack of focus, not knowing what I wanted to do with my life, and just plain being stupid and lazy. Several times I tried to "get it together" but by then it was too late as my grades were ridiculously low (mostly around 0) and I could never have gotten into grad school anyway.

Eventually I dropped out as I realized I wasn't getting anywhere. I had a lot of awards from things like programming competitions, so I used that to get some decent jobs. My intention was that I would work for a few years, clear my head, save up some money, and go back later.

Well, later is here. After 5-6 years work experience doing mind-numbingly boring menial programming jobs, I applied for and got accepted into a master's program at a fairly respected university. It's only a 1-year program, but if it goes well I'd definitely consider applying for a PhD.

Obviously it's very important to me that I do things right this time around, so my main questions are:

• Will the total lack of a BSc degree be a problem when applying for a PhD program?
• I spent the last half a decade mostly doing tasks which required no creativity or deep thought. Sometimes I feel like my mind is like an ex-boxer who has been sitting on a sofa eating KFC and donuts all day for 5 years.. completely out of shape! How do I get the gears spinning smoothly again?
• I need a plan for making the transition into "research" mode. To be honest when I was at university I never paid attention to what any of my friends were doing when they were reading papers so I have no idea how the whole thing works. e.g. Do I start by making friends with some professors? Should I volunteer to be a TA?
• At what point will I be able to apply for a PhD?

---

Master's degree will be in Computer Science.

Several people have expressed doubts about the legitimacy of a degree program that gives you a 1-year master's degree with no undergraduate degree. I just want to clarify that it's not some "IT shop under the railway bridge" giving the degree, it is a very respected university. It's a 1-year degree because masters programs are generally one year in the UK. Also, I didn't just walk in with nothing. I demonstrated in my interview that I had knowledge in CS topics, I had a couple published papers, as well as half a dozen international awards from my time as an undergrad, and work experience from top-tier Fortune 100 companies; I just never finished my degree.

Answer 1

As a current PhD student, my advice is to first ask yourself several questions before you decide that you want to apply to do a PhD.

First, why do you want to apply for a PhD program?

For some, the PhD is a stepping stone to a tenure-track job in academia. For others, the PhD is seen as a mind-enriching experience, even though the student wants to work in industry. For some industrial research jobs, having a PhD in a related field is a requirement before they will even consider hiring you for that job.

Second, what level of PhD program are you willing to accept?

There are also different levels of PhD programs. If you want to get into a good PhD program (MIT, CMU, Stanford, etc), you could take a look at current students of those programs and look at their CVs. Usually, these students would have graduated from a top school with a good GPA, and have had some undergrad research experience and maybe even a publication or two. If you want to get a job in academia, you usually would have to graduate from a good PhD program. If you only want to do a PhD to get a job in academia, it may not be worth your while to get a PhD from a less reputable school. However, this is a personal decision based on what is important to you.

Finally, are you both competent and interested in doing research in computer science?

The way that most people realize that they want to do research is that they have some sort of undergraduate research experience, or maybe even do a Master's.

Conclusion

If you think that you would want to get into a good PhD program in computer science, you should first get a BSc in a good school, and try to obtain some research experience while you are studying in order to get a feel for what research is like and whether you are able to do it and whether you really like it. I think that getting a BSc in a good school would probably be helpful in your career development even if you decide not to get a PhD.

At the end of the day, you make your own decisions and deal with the consequences. Good luck!

In response to OP's edits

A couple of thoughts:

• You didn't mention in your original question that you had several published papers. If you have some research experience and published papers, that definitely helps your PhD application.
• I don't have personal experience being on an admission committee. If you were able to meet someone on the admission committee of a university PhD program which you were interested in, you could ask him/her how they would evaluate your application. Since it is very rare for a PhD student not to have a BSc, I don't know how open they would be to accepting you as a student.
• Could you go back to the university which you dropped out of and finish of your degree in a year or two, by just completing missing requirements?
• From what I understand, most master's degrees are based on course-work, and have no research component. In that case, doing an MSc may not prepare you to do research as a PhD student. On the other hand, if you were able to do a perhaps 2-year MSc which includes research with an adviser/supervisor, this could be a great stepping stone into a good PhD program.

Answer 2

Will the total lack of a BSc degree be a problem when applying for a PhD program?

If the institution that has accepted you for the MS also offers the Ph.D., then if you do well in their master's program, you can almost certainly get into their Ph.D. program. If that seems to be a likely path, ask about their Ph.D. program and about how their Ph.D. graduates have done in academia.

If the Ph.D. is not available at your master's institution, then ask about their master's students who have gone onward to the Ph.D. and where they went. Ask a couple of those institutions whether your lack of the B.Sc. would be an absolute barrier to acceptance into their Ph.D. program assuming you've done well in your master's program. (Nobody is going to tell you whether they will admit you until you have that master's degree, but they probably will tell you whether the lack of the B.Sc. is an insurmountable barrier, which is what you need to know at the moment.)

I need a plan for making the transition into "research" mode

Talk to the people who put your name on their paper about how they came to the particular research they did and the process of completing it. Even if bioinformatics is not your field, the process is very similar for large areas of academic research. Some professors in Ph.D. programs publish lists of potential doctoral research topics. If you have an area of computer science in mind for your Ph.D. research, find professors who have published in that area and check their web pages. (Google Scholar is your friend.)

Do I start by making friends with some professors?

Even though you may be the same age as some of your professors, they're likely to want to keep a certain distance between themselves and their students. The advice already given is extremely good; that invitation to a social event isn't likely to be helpful.

If, in your career as a master's student, you come upon a professor whose research rings your chimes, and you have a cordial relationship with that professor, ask whether there's anything you can do to help with the research. If there is anything, expect it to be scut work, like programming others' algorithms, at least at first. Do not be dismayed; do the best job you can.

My own experience is similar to yours, but in the United States. I made a false start at an undergraduate education, worked in industry, finally completed the BS and MS degrees, and later the Ph.D. I have an academic job, and expect it to be the career from which I retire.

I don't know whether this is true in the U.K., but in the U.S. the master's degree qualifies one to teach in community colleges. Some universities, often lower-ranked, will also hire faculty with only the master's degree. That's how I got my start in academia. Check the credentials of the faculty at the institution that's accepted you for people without that Ph.D.

Answer 3

Will the total lack of a BSc degree be a problem when applying for a PhD program?

Chances are that it will. Your situation is uncommon, you will need to convince that your industry experience and your 1-year Master program* are somehow equivalent to a full undergraduate degree. You will be in competition with younger candidates who have a more traditional background, and a good grade record. This is not at your advantage. Note that some graduate schools/universities will simply not consider your application without an undergraduate degree, even if you found a professor willing to mentor you.

I spent the last half a decade mostly doing tasks which required no creativity or deep thought. Sometimes I feel like my mind is like an ex-boxer who has been sitting on a sofa eating KFC and donuts all day for 5 years.. completely out of shape! How do I get the gears spinning smoothly again?

You might have a too critical view of your work, it's unlikely that your everyday coding routine didn't include a bit of creativity. Even if you didn't recognize it as such. Plus, you probably have been in contact with software or techniques that are original.

But you're gym analogy is correct, practice is important. I would suggest to read papers (start with the most cited or most downloaded ones of the reputable journals in your field). Try to theorize the algorithm you would write to address or compete with the one presented, and maybe try to code some prototypes.

I need a plan for making the transition into "research" mode.

Have a look at these: Why do many talented scientists write horrible software? or Best-practice models for "research" code? They are full of excellent information about the difference between coding a commercial product and being a computer scientist.

Do I start by making friends with some professors?

That's not how you start, but it might happened along the way. More seriously, meeting professors during your Masters is an excellent idea. You will need to understand who's doing what and how you could be a part of it. If a professor knows you and is interested in your skills, it's going to be helpful, but it might not be enough (see above).

Should I volunteer to be a TA?

Teaching and administrative work will come by themselves soon enough, should you be accepted in a graduate program.

At what point will I be able to apply for a PhD?

That depends of many factors, but having some sort of degree is certainly the first step.

_{*In fact, an institution that gives Master degrees, in 1 year, to people without a bachelor would probably not be highly regarded by reputable graduate programs.}

Answer 4

Why don't you simply start with BSc studies?

If you know the stuff, you can do it quickly. Don't waste time with the lectures — just self-study and take exams. It does not need to take that long.

On the other hand, if you don't know it, then it will be very important to study it properly before starting your MSc & PhD studies.

Answer 5

I am going to make another attempt to answer the question.

The first question which you need to ask yourself is: Why do I want to do a PhD?

• If you want to do a PhD because you want to be a CS researcher, then you should aim to be admitted to a PhD program in a good CS department (eg top 10-15).
• If you want to do a PhD because you want to be a CS lecturer in a teaching university, it is possible to be employed at this job with only a MSc so getting a PhD may even be a waste of time.
• If you want to do a PhD because you want to get an industry job, then it is not vital to be admitted to a PhD program in a top CS department—any reasonably reputable CS department would suffice.

If you have decided that you want to get a PhD, then you probably want to get in the best CS department that you are able to. Thus, it is natural to ask the next question: How do I make a strong PhD application?

Chris Blattman says,

In short, focus on getting good recommendations, experience, grades and GRE scores.

You asked the question: Why should I spend 4 years to do a BSc, rather than spend 1 year to get an MSc?

• If you just want to get into a reasonable PhD program, and you are qualified enough to do so, then I agree that getting a BSc is a waste of time.

• If however you have decided that you want to get into a PhD program at a good CS department, then it may make sense to get your BSc. Why?

  • You could overload (do more courses in each semester) and get your BSc in fewer than four years. Anecdotally, I have heard of students at MIT finishing in three years for example.
  • Even though you are confident that you know everything that there is to know in an undergraduate CS education, unless you have the grades to back it up, PhD admission committees would be naive to just take your word for it. Earning a BSc is a certification that you are competent.
  • As part of your undergrad education, you have the opportunity to be involved in undergrad research programs. These programs are great in giving you the chance to work with professors and/or PhD students, for you to get a sense of whether you really like research, and for them to get to know you so that they can write you an outstanding recommendation letter, and if things go well for you to publish research paper(s).

Finally, I realized as I answered this question that Chris Blattman has a very long and detailed post that is a superset of my advice. I recommend that you take a look.

Good luck!

Answer 6

Another concern for others similarly situated that is not mentioned in any of the answers is that admission to a PhD requires meeting at least three criteria:

1. The right to study in a given country
2. Selection by the department
3. Approval by the university of the department's selections.

University approval may or may not be pro-forma depending on the institution, but it is possible that a university regulation might prevent the selection of a student without an undergraduate degree.

The right to study in a given country can also present a barrier. Obviously if you are a citizen, you don't need to worry, but many countries have requirements for issuing visas that might make this prohibitive. I have a Japanese friend who tried to study abroad in Norway, but he was unable to do so because he lacked the necessary number of years of education back in Japan. (14 rather than 12 were necessary and he similarly had not graduated undergraduate).

Answer 7

I'm a bit surprised that a good school accepted you into an MS program, much less a 1-year MS program, without a bachelor's degree. These days, some schools use their MS program as a cash cow, and the value of such programs is limited, in my opinion.

Regarding your specific questions:

Will the total lack of a BSc degree be a problem when applying for a PhD program?

Not necessarily ... if you do interesting work in your MS. This will require not only good grades but, ideally, an MS thesis or very serious project, even more ideally ending up in a publication. The problem is that with a 1-year MS, you will have no time to do anything but take required courses, and good grades in courses plus a mediocre industry experience and no BS will not make you a very appealing PhD candidate.

I spent the last half a decade mostly doing tasks which required no creativity or deep thought...How do I get the gears spinning smoothly again?

Do an MS thesis as part of your degree. Write some papers and impress your advisor.

Do I start by making friends with some professors? Should I volunteer to be a TA? At what point will I be able to apply for a PhD?

You should make professional friends with some professors (taking them to a ball game will not get you far). There are several ways to do this: take a particularly challenging course and be the top student or even less than the top student but demonstrate some unusual positive quality; do an interesting thesis that results in publications in respected venues.

Volunteering as a TA will help you expand your knowledge and experience with course material, but it will not, in itself, do much to help you get into a PhD program. PhD programs at good schools are extremely competitive, and to get into one you have to demonstrate skills that others do not have.

Incidentally, I would qualify your ability to survive at a mind-numbing job for 5-6 years as one of those skills. In research, perseverance is no less important than intelligence - I've seen very bright students leave a PhD program because their research wasn't advancing and they were sick of trying.

Answer 8

On some of the Ph. D. Applications I have seen, not only did they require a BSc. degree, it was required to have a 3.0 or above cumulative GPA in the undergrad.