The non-traveling mathematician problem This is a career question. I have just begun a research postdoc position in Southern California. It has been hard, but I've enjoyed teaching my first graduate courses and working on research and publishing.
From talking to other mathematicians, I've realized more and more that traveling to conferences a lot is an important part of being a research mathematician. But I don't want a job where I have to be gone from my wife and children on a regular basis. So research positions seem out of the question. 
So what are my options? I learned the hard way applying for jobs last year that companies aren't looking for bright young mathematicians who could learn stats, programming, etc. quickly when there are people with bachelors and masters in these areas who already have real-life experience. I know how to program in a few languages.
Thus, the clearest option seems to be a university position with a focus on teaching over research. But, it may be that such schools also would require regular travel. That's one reason I'm asking this question.
So,
1. What are the best options for a job not requiring travel given that I am a mathematician (in a non-applied area) with a great teaching record, an okay publishing record, and some computer science background, and
2. What can I do now so that I can get such a job when I apply in a year or two? No matter what, I'll work on publishing and teaching now because that's why I was hired; but what else can I work on?
Thank you your help!
 A: Have you ever considered the NSA or other government contractors? The standard way to get a job as a mathematician at the NSA is to apply for one of their development programs, which lets you tour around through several different groups within the NSA (but all would be in Maryland, probably) for 3 years and then settle down into one you like. They love math phds with some computer science background, and they pay more for those with a phd than for those coming out of bachelor's (so you wouldn't have "wasted your time" getting the degree; it would show up in your salary and promotion prospects).
Note that in order for this to work for you, you have to be a US citizen and you have to be able to pass a fairly invasive background check to get a top secret clearance. From when you apply to when you start working there might be 9 or 10 months. The development programs are called things like "The Applied Math Program" but I don't believe the math is any more applied than graph theory (though I can't know for sure). I know several pure mathematicians who are perfectly happy there. They don't get to research whatever they like, but they do get to choose problems from a list of problems and they seem to be pretty interested in what they do.
The other benefit of the government route is that there are contractors all over the place you could also go work for. There's even one in San Diego. It's called IDA (Institute for Defense Analyses). You can apply for those at the same time as you're applying for the NSA, or you could apply after the development program was done. Tread carefully here, I imagine the NSA doesn't like to lose people they've trained to other agencies, any more than any job would like losing people they've trained to a higher paying job. I only mention this in case living in CA is important to you.
Good luck!
A: Teaching/research job in any University (in a research oriented department or teaching-oriented
department) DOES NOT require a lot of travel. Invitations to conferences or seminar talks come
indeed but this does not mean you have to go, if you don't want to.
Traveling to two conferences per year (usually one week or less) and/or 1-2 seminar talks per year
(for one day) will be sufficient. Even if you do not travel at all, this is not going to hurt
seriously your research career. People travel a lot because they like to travel:-)
