What Do You Do with a M.S. in Systems Science?
Marianne reminded me of the girls I could never talk to in undergrad: former sorority president at Vanderbilt, self-confident, sharp, stunning, one of those people who make friends within seconds of walking into the room. I reminded Marianne of the young academics she grew up with: an unkempt, inarticulate super-calculator who was still in the process of learning how the numbers whizzing through his head could interact with the world, and completely at sea at times where the numbers did not apply. In a larger social group than grad school, we'd never have even known each other's names. But as two of the five or so Americans in a class of thirty, there was no reason for us not to be friends, and so we were. Marianne had spent a few years doing economic consulting in Boston, and despite all her social skills and natural talent she had hated it. So when I told her that I was leaving Northwestern to take a job in industry*, she smiled and said (kindly, but with her usual matter-of-factness) "You'll be back."
Marianne is usually right about these things, but the bridges were burning as I spoke, so I had to try. Getting a job in economics was right out; my engineering degree was so much more marketable (on paper) that my department was amazed that I was willing to pursue economics at all**. Getting a job in Systems Science should have been easy, with only one minor snag. I had to explain what it was that I did to the right person.
(Note to those of you who wish I would update more: this aside marks the one-hour mark since I started this post. And this one's going *fast*. That's why I don't update.)
Systems Science as a field (meaning generalized systems, not computer systems) is about fifty years old and not very big. There is no standardized notation or standard textbook. There is not even an agreement on what to call it. "Systems Science" is recognized mostly by the Aerospace industry. "Industrial Engineering" is widely known, but refers only to a subset of what I can do, and not my strongest suit at that. "Operations Research" is what it goes by in Defense. Why not just describe my skill set, you ask? Because few people would believe me. Basically, a Systems Scientist:
-Is called in to solve practically any problem
-Works with you to help you express your goals and possible courses of action mathematically
-Figures out how the parts of the problem interact and estimates the impact of each course of action
-Solves the problem directly if possible, by simulation if necessary
-Time permitting, leaves you with a program that can re-solve the problem if parameters change
It's the 'practically any problem' that trips up the people in HR. "So what's your specialty? Do you have experience in _____?" "My specialty is solving problems. I don't know anything about _____ yet, but I can learn everything I'll need to know in a couple of days." "What kind of programming languages do you know?" "Effectively none, and I haven't coded for years. But in a couple more days I can learn the language you want it in***." "...we'll call you."
Eventually, in my desperation, I started looking for "Modeling and Simulation" jobs. I make models. I build simulations. I should be able to hack it in a job like that. Eventually I was called back by a man from MITRE.
"Well, you would certainly do fine in our department, E526," he told me "but I think you belong in E525."
E525 was the Operations Research and Systems Analysis department. There was no hint on the MITRE webpage that they even existed, and they had no job openings at that time. They flew me out anyway, and I gave them one of the worst presentations I've ever given and left behind a folder containing all the receipts I would need to be reimbursed for several hundred dollars of hotel and rental car costs.
They hired me anyway. I've been there a month. In that time, I've worked on problems ranging from early detection of epidemics to combinatorial auctions to *CLASSIFIED*. The pay is good, the benefits are great, the people are top-notch and I can't be asked to work more than forty hours a week. Turnover is incredibly low; people leave for other departments or other sites, but rarely for other companies and almost almost voluntarily.
Sorry, Marianne. I love it here. I might come back someday, but it'll be long after you're a department head.
(Two hours of writing. Gah!)
* Something you all might have wanted to know too, I bet. I'm sorry. This was a very stressful time in my life, and the feelings running through my brain were not conducive to blogging.
** Oh, Economics, how I love the discipline I thought you were!
*** Not joking. A week after first seeing Python, I used it improve the run time of a program from twenty minutes to ten seconds.
Marianne is usually right about these things, but the bridges were burning as I spoke, so I had to try. Getting a job in economics was right out; my engineering degree was so much more marketable (on paper) that my department was amazed that I was willing to pursue economics at all**. Getting a job in Systems Science should have been easy, with only one minor snag. I had to explain what it was that I did to the right person.
(Note to those of you who wish I would update more: this aside marks the one-hour mark since I started this post. And this one's going *fast*. That's why I don't update.)
Systems Science as a field (meaning generalized systems, not computer systems) is about fifty years old and not very big. There is no standardized notation or standard textbook. There is not even an agreement on what to call it. "Systems Science" is recognized mostly by the Aerospace industry. "Industrial Engineering" is widely known, but refers only to a subset of what I can do, and not my strongest suit at that. "Operations Research" is what it goes by in Defense. Why not just describe my skill set, you ask? Because few people would believe me. Basically, a Systems Scientist:
-Is called in to solve practically any problem
-Works with you to help you express your goals and possible courses of action mathematically
-Figures out how the parts of the problem interact and estimates the impact of each course of action
-Solves the problem directly if possible, by simulation if necessary
-Time permitting, leaves you with a program that can re-solve the problem if parameters change
It's the 'practically any problem' that trips up the people in HR. "So what's your specialty? Do you have experience in _____?" "My specialty is solving problems. I don't know anything about _____ yet, but I can learn everything I'll need to know in a couple of days." "What kind of programming languages do you know?" "Effectively none, and I haven't coded for years. But in a couple more days I can learn the language you want it in***." "...we'll call you."
Eventually, in my desperation, I started looking for "Modeling and Simulation" jobs. I make models. I build simulations. I should be able to hack it in a job like that. Eventually I was called back by a man from MITRE.
"Well, you would certainly do fine in our department, E526," he told me "but I think you belong in E525."
E525 was the Operations Research and Systems Analysis department. There was no hint on the MITRE webpage that they even existed, and they had no job openings at that time. They flew me out anyway, and I gave them one of the worst presentations I've ever given and left behind a folder containing all the receipts I would need to be reimbursed for several hundred dollars of hotel and rental car costs.
They hired me anyway. I've been there a month. In that time, I've worked on problems ranging from early detection of epidemics to combinatorial auctions to *CLASSIFIED*. The pay is good, the benefits are great, the people are top-notch and I can't be asked to work more than forty hours a week. Turnover is incredibly low; people leave for other departments or other sites, but rarely for other companies and almost almost voluntarily.
Sorry, Marianne. I love it here. I might come back someday, but it'll be long after you're a department head.
(Two hours of writing. Gah!)
* Something you all might have wanted to know too, I bet. I'm sorry. This was a very stressful time in my life, and the feelings running through my brain were not conducive to blogging.
** Oh, Economics, how I love the discipline I thought you were!
*** Not joking. A week after first seeing Python, I used it improve the run time of a program from twenty minutes to ten seconds.