How can I get research experience outside the college setting?

Question

I graduated 4 years ago from a small college with a good computer science program, and I am interested in going back to grad school. My numbers (GPA & GRE) are good, but those don't matter nearly as much as I wish they did!

CS grad programs understandably value research and recommendations in an application above all else, and I have none. I worked every summer when I was an undergraduate and didn't do any student research. I can't work it into my job, as we are a very fast-paced company and all available time is spent directly developing on our products. I would like to be able to continue working while doing this, though that may be tricky.

How can I build research experience in a way that would be attractive to future grad schools now that I am out of college and don't work for a research lab?

Answer 1

Most likely, if your GPA and GRE scores are fairly well you will not have a problem getting into a graduate program. I was in a situation very similar to yours, and I had no problem getting into a program. Most programs will value your work experience. Just be sure to get at least one recommendation letter from a person who knows your academic strengths. For that, I just got a couple of my former professors to give me a recommendation.

Good luck to you!!

Answer 2

As far as recommendations are concerned, you do not need letters from big names.

You need to get letters teachers/professors who knew/know you well, convey to the reader that they knew/know you well and can communicate effectively his/her strong convictions about your potential.

Answer 3

I was able to get into a good school without much research experience (and several years of work experience), but it also turned out that I am not a great researcher (better at implementing other's research). Fortunately, Computer Science is an area where for a number of disciplines, you do not need a research lab to do at least a small research project in your personal time. I would suspect that would help along with getting the appropriate references (which I did have as well). Your former teachers should be willing to give you recommendations as well as your current company.

Also, one of my former employers, even though a startup, gave me some free time to finish my PhD, so you would be surprised what you can get if you ask (although you know your company better than I).

Answer 4

You could work on little demo or open-source projects in your field. Your contributions will help

1. expose you to the field, giving you experience
2. determine if this is what you actually want to dedicate your life to

In my field (scientific visualization), I have developed demo programs in VTK visualizing publicly-available data sets. These side projects weren't ground-breaking, but they taught me new things and demonstrated my ability to learn things on my own... an important skill for a graduate student.

Answer 5

Sourceforge has a number of research projects - you could work on those.

You might also try dropping into your local computer science department and see if they would like some help on one of their projects.

Answer 6

Well... you could help me in a project I am currently working on, to get some visibility.

In any case, research is a particular and complex world. The best strategy to do research is to get new ideas. To get new ideas, you don't wake up one morning with a new idea. You actually take ideas from others, and add a small bit. Newton said "if I've seen that far is because I was standing on the shoulders of giants". To do this, you actually have to know your "giants", before climbing them.

As a result, my suggestion is to read scientific papers of an interesting discipline you want to contribute on, possibly the same of a research group you have close access to. Don't worry if you don't grasp the paper. You probably will not get it immediately. You have to grow, understand. Perform a full bibliography search of the argument: look the citations, and get to know what's going on in the field relative to that problem, since the origins. Try to see open issues, interesting questions or problems in the current models you can contribute.

Try to build some kind of contact with a professor in the field or anyone under his supervision, by asking papers, for example. Don't bother him too much. They are very busy people. Go to a local university library to access the scientific papers, try to be present and visible somehow. When you have an idea, propose it to your favourite professor or postdoc (they are normally more interested in new projects, as their contracts are short and their stay depends on new things to do). If you get to have a chat, and get to have a project, you will probably end up with a completely different direction, idea or task than your initial one.

Continue this way. When an opening appears, you will be skilled enough to propose yourself and join the group without much hassle.

As you can see, it's mostly social networking, and it takes a lot of time and effort. I doubt you can do this while keeping your job. One fundamental recommendation: be careful of who you pick. Check how well funded s/he is, if s/he does active research, and how large is the group. Remember that postdocs and Ph.D. will be your main source of knowledge. Professors are too busy.

Answer 7

Two important “tasks” of grad school/research are : writing research papers and teaching.

Research papers : Choose a topic you like/know a lot about/etc., read the latest literature on the subject and try to write 2 or 3 original papers on the subject (and if you think they are worthy of publication, go for it). A “State of the art” paper on a relatively new field is also a possibility.

Teaching : Choose a classic subject in computer science (data structure, algorithmic, formal language, graph theory, etc..) and write a textbook for undergraduates.

Add to the mix a “real” programming experience/project, and you’ll have a nice portfolio to attach to your statement of purpose.

Answer 8

Well, if you are young enough to qualify, graduate exchange programs might get you a foot in, on another country.

Another option, and one you should consider seriously, even when looking at exchange programs, is to find researchers in the field you like, and contact them. Recommendations goes a long way, but having an interested and dedicated student is something any research would like. Showing initiative and interest might well offset your low points.

And, at any rate, getting to know researchers in the field you like -- and their work, of course -- will be useful down the road anyway, so why not do it now?

Answer 9

I'm in the same situation.

In addition to phd students and post-docs, seek out visiting professors. They often don't have their own grad students and need help with their research.

Also, try to figure out what you are interested in. CS is a big field, and research in theoretical sub-fields (computation, crypto) is quite different from research in the more applied sub-fields (OS, security).

Answer 10

I don't know if it's viable and it's definitely not cheap, but you can try to ask your employer for a year off and enroll in a MSc., so that you can put a foot back in some university.