So my long job search is finally over, and I got a job at Numerica Corporation. I had interviews at a couple interesting places. One was Two Sigma Investments, a financial company based in New York City. I got there a day before the interview, which gave me some time to explore the area. I tried to get a taxi to the Dave & Busters in Times Square, but I tried to get in a taxi and they said they didn't know where it was (even after I showed them the map on my iPhone.) I took a subway there but when I tried to go back, my phone was out of battery power so I couldn't use the GPS to find my way back. I again had no luck finding a cab, so eventually I had to go into a subway station and use the pay phone to call the hotel to get directions back. The next day I had my interview which ended early, so I got to use the extra time to go to a local game store. Then I got to the airport and the flight was delayed. The flight landed back in Chicago about two hours later, and I had assumed I had missed my connection so I used the rebooking phone to call for rebooking, and they said that the only flights they had were for two days from then. I decided instead to just stay in a hotel and take the train home the next morning, but while I was walking back to talk to the airline about reclaiming their backs I saw the departure board which saw the connecting flight was also delayed, so I was able to get on it and get home. Also, one of my professors was on that plane coming back from an NSF panel, and he took me home. Unfortunately I didn't end up getting the job.
Lots of interesting stuff happened on the interview for Numerica. My flight was originally scheduled to leave at 7:00 AM. I overslept and woke up at 6:30, and rushed out the door in the vain hope of getting to the airport in time. I got there at about 7:10 but fortunately the flight was delayed until that afternoon. I got rebooked on a 2:15 PM flight which gave me time to go back home, do some stuff at school that I needed to do, get the charge for my laptop (which I left at home), then go back to the airport. On the flight from Champaign to Chicago a different professor I know was also on this flight. I landed at the Denver airport and took a taxi out to Loveland. The hotel was next to a strip mall so I was able to go out to dinner and also finish up a presentation on my research, which I gave as part of the interview. The interview went vrey well. During lunch, I mentioned that I was interested in board games, and one of the people who worked there was also interested in board games and knew of some board game stores in the area, and offered to take me to them that evening. It was probably the best interview experience I have ever had.
So I decided to accept their offer, and I will be starting work by the end of August. I definitely think that stopping with a Masters degree rather than continuing with a Ph.D. was the right thing to do. I'll be able to continue doing what I love doing, and I'll also be making a lot more money than I am as a graduate student. I am not sure what I will do with all that money, but there are other people in this world who need the money a lot more than I do.
Showing posts with label Jobs. Show all posts
Showing posts with label Jobs. Show all posts
Tuesday, May 10, 2011
Wednesday, March 16, 2011
Job Search
So, last week I made my decision: I will be switching to a Masters degree program. This does mean that I won't get to do that internship at Lawrence Livermore, because the internship requires that you commit to staying in school through the Fall semester. This actually turns out to be okay because I will get to stay at school through August to finish up my thesis. There's a fair bit of work still left; I have to finish up all the remaining programming work, write documentation on all my code so that my replacement can figure out what's going on after I leave, and of course write the actual thesis. Also, I have started to look for full-time jobs for after school.
I went to the Engineering Career Services office and got my resume and cover letter looked at, and they basically had me rewrite the whole thing using the format described in the career guide, which made it look much better. I took the new resume and cover letter over to the Graduate College's career services office, and they told me that my resume and cover letter looked very good, and they only had a couple minor stylistic changes. I searched for jobs online and through job posting on bulletin boards in the Computer Science department. So far I have submitted resumes to 9 companies, and have gotten phone interview requests from 2 of them. One of them was MathWorks, which makes Matlab. Another was Palantir Technologies, which makes a data analysis and visualization platform (read the web site if you are interested). They wanted to interview me for a "Business Development" position, and I'm very interested in what that position entails. I'll report back when I see how the interviews go.
And one more thing. When I was in elementary school and people asked me what I wanted to be when I grew up, I would answer "math test-question editor", because I took lots of math tests like the Math Olympiad during that time and often I found the test questions ambiguous. During my online job search, I found an online job advertisement for a math test-question editor. But things have changed a lot in the past 15 years, and I decided not to apply for that job.
I went to the Engineering Career Services office and got my resume and cover letter looked at, and they basically had me rewrite the whole thing using the format described in the career guide, which made it look much better. I took the new resume and cover letter over to the Graduate College's career services office, and they told me that my resume and cover letter looked very good, and they only had a couple minor stylistic changes. I searched for jobs online and through job posting on bulletin boards in the Computer Science department. So far I have submitted resumes to 9 companies, and have gotten phone interview requests from 2 of them. One of them was MathWorks, which makes Matlab. Another was Palantir Technologies, which makes a data analysis and visualization platform (read the web site if you are interested). They wanted to interview me for a "Business Development" position, and I'm very interested in what that position entails. I'll report back when I see how the interviews go.
And one more thing. When I was in elementary school and people asked me what I wanted to be when I grew up, I would answer "math test-question editor", because I took lots of math tests like the Math Olympiad during that time and often I found the test questions ambiguous. During my online job search, I found an online job advertisement for a math test-question editor. But things have changed a lot in the past 15 years, and I decided not to apply for that job.
Friday, February 6, 2009
More on the Census Bureau
The problem I am working on for the Census Bureau is known as "unbiased controlled rounding." The problem is the following:
GIVEN:
- A two-dimensional array of values, each in the range [0,1).
- The marginal counts - i.e. the sum of each row, and the sum of each column (and the grand total.) We know that the row sums and column sums are all integers, and they of course are equal to what the values in each column add up to.
FIND:
- A randomized method of producing an array of zeros and ones the same size as the original array such that:
-- The expected value in each cell is the same as the corresponding value in the original array.
-- The row sums and column sums are the same as those in the original array (all the time, not just on the average.)
---
The main application of this problem is to reduce the risk of disclosure when releasing "microdata" - data with very small sample sizes (for example, all the businesses within a particular area that's ten blocks wide) Basically you want to round off the answers that the respondents gave so that an adversary cannot identify an individual respondent by their answers, but still preserve all the aggregate statistical properties (like mean, variance etc.) and avoid introducing bias.
As it turns out, the two-dimensional case has a known solution. The three-dimensional case (where in addition to row and column sums, you have sums going in the third direction, and you also have sums through planar "slices" of the cubical "array") does not, and in fact it is not always possible to solve.However, the goal is to get as close as possible to a solution (say, make sure that the marginal counts are off by at most 2) and do this in a reasonably computationally efficient way.
---------------
BONUS QUESTION: Solve the problem given above for the one-dimensional case (i.e., when the array just has one row in it, and the only marginal sum there is is the grand total). The solution to this problem gives an idea of some of the techniques used in higher dimensional solutions.
GIVEN:
- A two-dimensional array of values, each in the range [0,1).
- The marginal counts - i.e. the sum of each row, and the sum of each column (and the grand total.) We know that the row sums and column sums are all integers, and they of course are equal to what the values in each column add up to.
FIND:
- A randomized method of producing an array of zeros and ones the same size as the original array such that:
-- The expected value in each cell is the same as the corresponding value in the original array.
-- The row sums and column sums are the same as those in the original array (all the time, not just on the average.)
---
The main application of this problem is to reduce the risk of disclosure when releasing "microdata" - data with very small sample sizes (for example, all the businesses within a particular area that's ten blocks wide) Basically you want to round off the answers that the respondents gave so that an adversary cannot identify an individual respondent by their answers, but still preserve all the aggregate statistical properties (like mean, variance etc.) and avoid introducing bias.
As it turns out, the two-dimensional case has a known solution. The three-dimensional case (where in addition to row and column sums, you have sums going in the third direction, and you also have sums through planar "slices" of the cubical "array") does not, and in fact it is not always possible to solve.However, the goal is to get as close as possible to a solution (say, make sure that the marginal counts are off by at most 2) and do this in a reasonably computationally efficient way.
---------------
BONUS QUESTION: Solve the problem given above for the one-dimensional case (i.e., when the array just has one row in it, and the only marginal sum there is is the grand total). The solution to this problem gives an idea of some of the techniques used in higher dimensional solutions.
Friday, January 23, 2009
First week at the Census Bureau
So today is the end of my first week at the U.S. Census Bureau.
Monday and Tuesday, the Census Bureau was closed because of Martin Luther King Day and the inauguration. (Although apparently I still get paid for those days!)
Wednesday and Thursday was orientation. Yes, it lasted two full 8-hour days. Most of it was pretty mundane presentations, a lot of which didn't even apply to me because I am a part-time worker (like the presentations about health benefits, retirement plans, etc.) However there were a couple funny parts, like the presentation about "Government Ethics." Normally, of course, you can teach government ethics just by opening up the newspaper to a random page. However, since this was the day after inauguration, all the newspapers were filled up with inauguration news, so we learned about government ethics by watching a video entitled "The Battle for Avery Mann," which features a nondescript government worker in a nondescript office facing a series of ethical conflicts (whether to use the office copying machine for personal use, whether to accept gifts from a subordinate, what to do when asked to review a proposal from a company he also works for part-time) and two comical characters representing his good and evil impulses trying to tell him what to do.
On Friday, I started doing some actual work. I am working in the Statistical Research Division in the Disclosure Avoidance Research Group. The goal of this research group is to find ways of releasing census data to the public in a way that is useful to potential users of the data, but that does not enable anyone to find out any information about a particular respondent. Methods that are being used now, or being explored for potential use in the future, for doing this include:
1. Providing only certain cross-tabulations of data, not the full data set.
2. Suppressing cells in cross-tabulations with less than a certain number of people in them.
3. Adding artificial "noise" to certain elements of the data set.
4. Providing a synthetic data set with similar statistical patterns to the real data set (say, the same values for all cross-tabulations with up to a certain number of variables) but that doesn't have any of the real data.
To see some of the work the Statistical Research Division does, you can see research papers they have published here. I am working with Yves Thibaudeau and Robert H. Creecy.
Bonus Math Question:
The Census Bureau offers a benefit known as a "health savings account" (HSA). Employees can designate a portion of their income to go into the HSA pre-tax (meaning that money put into the HSA is deducted from your income for purposes of income tax). Money in the HSA can be used for health care expenses. However, any money left in the HSA that is not used for health-care expenses is lost.
How can you determine the optimal amount of money to put in the HSA? (Assume that you are risk-neutral, that your utility of money does not depend on your health care expenses, and that you know what your probability distribution of health care expenses for the next year looks like.)
Monday and Tuesday, the Census Bureau was closed because of Martin Luther King Day and the inauguration. (Although apparently I still get paid for those days!)
Wednesday and Thursday was orientation. Yes, it lasted two full 8-hour days. Most of it was pretty mundane presentations, a lot of which didn't even apply to me because I am a part-time worker (like the presentations about health benefits, retirement plans, etc.) However there were a couple funny parts, like the presentation about "Government Ethics." Normally, of course, you can teach government ethics just by opening up the newspaper to a random page. However, since this was the day after inauguration, all the newspapers were filled up with inauguration news, so we learned about government ethics by watching a video entitled "The Battle for Avery Mann," which features a nondescript government worker in a nondescript office facing a series of ethical conflicts (whether to use the office copying machine for personal use, whether to accept gifts from a subordinate, what to do when asked to review a proposal from a company he also works for part-time) and two comical characters representing his good and evil impulses trying to tell him what to do.
On Friday, I started doing some actual work. I am working in the Statistical Research Division in the Disclosure Avoidance Research Group. The goal of this research group is to find ways of releasing census data to the public in a way that is useful to potential users of the data, but that does not enable anyone to find out any information about a particular respondent. Methods that are being used now, or being explored for potential use in the future, for doing this include:
1. Providing only certain cross-tabulations of data, not the full data set.
2. Suppressing cells in cross-tabulations with less than a certain number of people in them.
3. Adding artificial "noise" to certain elements of the data set.
4. Providing a synthetic data set with similar statistical patterns to the real data set (say, the same values for all cross-tabulations with up to a certain number of variables) but that doesn't have any of the real data.
To see some of the work the Statistical Research Division does, you can see research papers they have published here. I am working with Yves Thibaudeau and Robert H. Creecy.
Bonus Math Question:
The Census Bureau offers a benefit known as a "health savings account" (HSA). Employees can designate a portion of their income to go into the HSA pre-tax (meaning that money put into the HSA is deducted from your income for purposes of income tax). Money in the HSA can be used for health care expenses. However, any money left in the HSA that is not used for health-care expenses is lost.
How can you determine the optimal amount of money to put in the HSA? (Assume that you are risk-neutral, that your utility of money does not depend on your health care expenses, and that you know what your probability distribution of health care expenses for the next year looks like.)
Saturday, December 13, 2008
Seduced by Amazons?
"Open the case."
-Howie Mandel
-----------------------------------------
I am currently in Seattle after having finished interviews with Amazon.com. I have already been offered a full-time position but went over to talk to them and find out whether this was someplace I want to work. Pertinent information I found out was as follows:
-Howie Mandel
-----------------------------------------
I am currently in Seattle after having finished interviews with Amazon.com. I have already been offered a full-time position but went over to talk to them and find out whether this was someplace I want to work. Pertinent information I found out was as follows:
- Amazon is divided up into several "organizations," and each organization has several "teams." I talked to two organizations (Distributed Systems Engineering and Financial Systems) and two to three teams within each organization. I found one team (Customer Experience Analytics, which analyzes data from how customers use Amazon.com to find out how to improve the Web site to generate more sales) that I want to work at, but I am not particularly interested in any of the other teams that I met with.
- It is not possible to lock in a position on a particular team when you choose to join. Instead, you select an organization. If you do not select an organization one will be selected for you. Then when you get closer to the start date (about a month or so out) you get to choose a team from all the teams that have openings. Of course the teams that have openings are constantly in flux, so there is no way to know what teams will be available when it comes time.
- The deadline date for accepting the offer is December 19. Thus I will have no idea which graduate schools I will get into when accepting this offer. I analogized this situation to "Deal or No Deal", because I have to decide whether to take the "Deal" at Amazon before knowing enough information about it.
- I talked to my recruiter about this situation and he said that one possibility is to decline this offer, and then in May 2009 call again to see what teams are available, and join a team if there's a team with openings then I want to be on. (they have already evaluated my qualifications so I don't have to go through the interview process again.) Of course there's always the chance that all the spots will be full by then. The way I think about it is that this option is a dominant strategy: if all the teams I want fill up by then, then I wouldn't want to work at Amazon anyway, while if not, then I can still get an offer.
I will probably ask my recruiter again about what happens if I sign the offer letter now, and then when the time comes find out that there aren't any teams I want to work on, but the above option is probably what I'm going to do. So in a few months, I'll find out what was in the "cases" that I opened. (I think one of the other students I was with, when I mentioned it being like "Deal or No Deal," said something about opening a door to find out what prize is behind it. I reminded him that doors as a device to hide prizes in game shows were outdated 20 years ago, and the only reason anyone remembers them is because of the associated math problem.)
Friday, December 5, 2008
What's this about a recession?
"Researchers at Harvard say that taking a power nap for an hour in the afternoon can totally refresh you. They say that by the time you wake up you'll feel so good, you'll be able to start looking for a new job."
- Jay Leno
---------------------------------------
I just learned that I have been offered a full-time job at Amazon.com as a Software Development Engineer for after I graduate. On December 11-14 I will be traveling over there in order to tour the campus, learn about what a great place Amazon is to work, and talk to different project teams to find out which ones I might want to work on.
I have also been offered an internship at the U.S. Census Bureau's Statistical Research Division to start sometime in January. There, I could quite possibly be working with some of the people responsible for the development of the "imputation" techniques that were the subject of a 2002 Supreme Court case.
- Jay Leno
---------------------------------------
I just learned that I have been offered a full-time job at Amazon.com as a Software Development Engineer for after I graduate. On December 11-14 I will be traveling over there in order to tour the campus, learn about what a great place Amazon is to work, and talk to different project teams to find out which ones I might want to work on.
I have also been offered an internship at the U.S. Census Bureau's Statistical Research Division to start sometime in January. There, I could quite possibly be working with some of the people responsible for the development of the "imputation" techniques that were the subject of a 2002 Supreme Court case.
Subscribe to:
Posts (Atom)