Monday, February 2, 2009

Hertz Fellowship Interview

Shortly after submitting my application for the Hertz Foundation Fellowship, I found out that I'd been selected for a first-round interview (of the approximately 800 applicants, 200 are chosen for a first-round interview, 50 for a second-round interview, and 15 for the award). I had heard the interview would be technical in nature and that it wasn't really possible to prepare for. I was worried that "technical" meant that I would be asked to apply Green's theorem and do problems on special relativity, so I tried to brush up a bit on math, physics, and electronics. The questions turned out to be much more general than I expected.

On the day of my interview, I put on my nicest clothes and biked over to the hotel I'd been told to go to. There were clearly a lot of interviews going on that day--when I showed up, several other students were milling around in the lobby waiting to be interviewed at the same time that I would be. I'd found out when I called the Foundation a few days earlier that one of my interviewers would be a member of the board of directors of the Foundation; my other interviewer would be a recent Hertz fellow (I was given their specific names, and looked them up beforehand; more on this below). I was quite intimidated and nervous. When the time came, I went up to what appeared to be my interviewer's hotel room (I had assumed the interview would be held in a conference room or something).

Both of my interviewers tried to come off as being "on my side" in the sense that they were very friendly and did not make me squirm (for too long) when they asked questions I couldn't answer. The point of the interview is to get at how an applicant thinks, so they don't like silence. At several points when I was trying to reason through something, my interviewers told me that I needed to think out loud and that silence would be counted against me (or something along those lines). I don't remember all of the questions they asked--in fact, I think I disproportionately remember the ones I had trouble with, since I spent more time on those--but here are the ones I remember. (At the end of the interview, I explicitly asked my interviewers if I could write about / share these questions and they said I could and claimed that they do not re-use questions.)

1. I was asked a seemingly simple question that I could not answer: If I weigh myself on three consecutive days, assuming that my weight fluctuates a little bit and does not stay the same, what is the chance that my weight on the 2nd day is between what it is on the 1st and 3rd days? I told them I thought the answer was 1/2 and why I thought this, and they told me I was incorrect. I tried thinking about it more and clearly wasn't getting anywhere, and they said we could come back to it later. Thankfully, we never did (this was how they addressed most of the questions I got stuck on--they said we'd move on to something else temporarily, but we never returned). I'm not sure if the point was to get me to confirm that my weight measurements were normally distributed, or something.

2. If I had an extremely sensitive scale and a mass, what factors would affect the weight registered on the scale? This question was pretty straightforward; I listed off things such as altitude, air pressure, etc. They prodded me towards a major one that I was missing (the relative location of the moon), and we moved on.

3. They moved on to more chemistry related stuff at this point. They asked me a few factual questions about how resonance works, and asked me if I could identify who first proposed the structure of benzene. They then asked me whether the resonance stabilization in benzene vs. static 1,3,5-cyclohexatriene can be predicted from ab initio calculations; I replied that it was easy to do with a pencil, paper, and MO theory.

4. Next, I got pretty stuck on a question that should have been much easier: If I have a cup of ice water, when the ice melts, will the level of the water go up, down, or stay the same? My initial response was that it would go down, assuming that the volume of ice sticking out above the surface was negligible. It was quickly pointed out to me that that was a faulty assumption. After at least five minutes of making an idiot of myself and after they reminded me about Archimedes's principle, I arrived at the answer. A friend of mine later pointed out to me that sea levels rising due to global warming was a good analogy for this. That would have been a perfect response, demonstrating the type of thinking that (I believe) they are looking for in these interviews.

5. I was asked a few questions about electrochemistry, which is an area where I have a real dearth of knowledge. I was asked to explain what happens to a battery when it is cooled, which went well. Then I was asked how a lead acid battery works; I had no clue. They talked me through that one.

6. Continuing in the battery vein, they asked me why lead acid batteries had been used in cars for such a long time. I thought they were looking for a definite answer to this and I wasn't sure what it was. I realized (a little too late, perhaps) that they were using this question to gauge my ability to talk about technical issues, and they were happy to hear my speculation if I presented it confidently and could back it up with solid reasoning. I mentioned hydrogen storage offhand and then they asked me about the state of the field, current barriers to adoption of hydrogen as a fuel, etc. So here our interaction was more of a conversation--they really just went off what I was saying, and if I had started talking about something I didn't know about, they would have found out very quickly.

7. Next, they asked me how a specific spectroscopic method employed in chemistry works. They would not have brought this up with most people, but it was directly related to something in my application. I stumbled a bit at first, since it had been a while since I thought about it, but this went okay.

8. Finally, I was asked some conversational questions about my intended field of study in graduate school. They didn't seem to have much knowledge of the field of chemistry that I want to work in, and asked me some basic technical questions about it (their questions were along the lines of asking a natural products chemist what was left to do aside from make increasingly-complex molecules, or asking a medicinal chemist about the drug design process), asked me to discuss its importance, why I was interested in it, etc.

9. They asked me if I had any questions and I told them I was planning on writing about my experience and asked if it was okay to share the questions they asked. They said they couldn't see why not. I would have asked another question or two, but our hour was up and it was time to part ways.

One thing that I realized during the interview was that while my interviewers had read my application, they did not remember it very well, since they had undoubtedly read dozens of applications. They had taken a few notes on it, jotting down a few key phrases to ask me about during the interview. I had Googled my interviewers beforehand and neither seemed to have formal training in chemistry, but I was surprised by the depth of the chemistry knowledge of the younger interviewer, given that he was a graduate student in computer science.

Along these lines, the application allows them to make the first cut, but once you're at the interview stage, it seems like the application is pretty much irrelevant. I got the sense that they would be deciding whether I advanced to the next round based entirely on my performance at the interview.

I think I was a little too flustered during the interview and made several errors because I felt pressure and stress. But I do now have a good sense of what they were looking for. They rushed through the questions that I answered correctly and the subjects that I knew well. The point of the interview was to see how I thought about problems, so when I didn't know the answers to their questions, my responses were far more instructive to them.

After the interview, I expected that I would not make it to the second round; today, I received an email saying I had not been chosen for further consideration. There's always next year, if I can actually muster the motivation to apply again.

8 comments:

  1. Good description. I would try again next year - what do you have to lose - especially now that you have an idea of what to expect.

    ReplyDelete
  2. For the first question, one of the three measurements is in the middle, so there is a 1/3 chance of any of the three being the middle measurement, discounting the possibility of obtaining the same value twice.

    ReplyDelete
    Replies
    1. The "1/3 chance" would be true if the three measurements were independently distributed. But your weight today is more strongly correlated with your weight yesterday than it is with your weight two days ago. So the 1/3 chance argument is wrong.

      If your weight is a C^1-continuous function of time and you measure it with a sufficiently small time step, then almost every measurement is between the measurements immediately preceding and following it. But as the time step increases to infinity, the measurements approach independence (never quite reaching it), so I think the probability would approach 1/3 (from above) as per your argument.

      Delete
    2. I estimate 1/4.

      For the third day's weight to be between weights 1 and 2, you need two things:
      1. Either your weight goes up on day 1 and down on day 2, or vice versa. Assuming your weight function is symmetric (ie on average you do not gain weight), the possible configurations are ++, +-, -+, --. Two of these (+- and -+) are allowed, so p=1/2.
      2. The weight change on the second day must be smaller than the weight change on the first day (else you overshoot). As Anonymous argued, the odds of one weight-change being smaller than the other are even, so p=1/2.

      (1/2)*(1/2)=(1/4)

      Delete
  3. when ice melts in water the water level stays the same...

    ReplyDelete
    Replies
    1. You are correct. Shows you what I know!

      Delete
  4. So do you think based on the questions during the interview, that if you didn't have a strong science/math background, that you wouldn't pass? I'm just wondering if I should apply. I only have some of the "core" science courses (not even physics!)

    ReplyDelete
    Replies
    1. It's hard for me to say given that I don't know about other people's interview experiences. But my guess is that if it's obvious that you don't have a strong background in the physical sciences but you are still eligible for the fellowship, they will take that into consideration. I think the point of the interview is to assess your problem solving skills when confronted with something new, and they should be able to do that within your field regardless.

      Delete