Graduate experience

This page goes into some detail regarding my experience as a graduate student at the University of Chicago. The page focused on my personal experience. For my write-up on the overall experience, see this Quora answer. For more links on what I did in graduate school, see my graduate studies page and my teaching page. Also relevant is my Quora answer to the question Why did you leave academia? (mirrored here in case you prefer not to access Quora).

As with my undergraduate experience, the most important aspects of my experience were different from those I was initially focused on acquiring. I am including most things that happened to me while in graduate school, even those that weren’t related to graduate studies, with the exception of some things I did near the very end where I was basically done with graduate school and just waiting to leave.

Quick chronology

While a chronological story isn’t the best way of describing my graduate school experience, outlining the approximate chronology provides more context to the full story. So here it is:

  • First year (September 2007 – August 2008): This year was mostly spent on graduate coursework (a total of nine courses, three per quarter for three quarters).
  • Second year (September 2008 – August 2009): This year, I was a College Fellow for the undergraduate algebra (honors) sequence. On the research side, I had to complete my topic proposal and finish up my language exam to get my Masters degree.
  • Remaining years (September 2009 – December 2013): On the research side, I worked on my thesis and other group theory research. I also began full-time teaching. On the personal and side project fronts, I laid the foundation for many things that would define my life and ultimately lead me to find a home for myself outside of academia.

My first experience with financial independence and significant income

Although I come from a fairly well-off family and my parents have been supportive of many of my spending choices, it’s a different feeling to have one’s own source of income that goes beyond one’s basic needs and gives one the flexibility to save, spend, or invest in a variety of ways. A graduate student salary isn’t a huge amount, but at the time it was a big change from my earlier life. I used to cook all my meals, and had a lot of mental and emotional space with few money-draining hobbies, plus I had a generous initial scholarship, so I quickly accumulated a small but reassuring financial buffer.

While this didn’t immediately change a lot for me it did affect the way I thought about and related with money.

Graduate school culture and community

Graduate school was a significantly different experience for me than undergraduate life. The University of Chicago was a much larger, wealthier, bigger institution than Chennai Mathematical Institute, where I did undergraduate studies. There were differences in wealth and lifestyle in general. We had a lot more homework and more demanding institutional requirements. Partly, this was due to the unusual rigor of the University of Chicago’s program. Partly this was because the workload in American universities was in general higher, with clearer and more definitive rules for homework. These differences weren’t greatly jarring, and I didn’t face significant difficulties on account of them.

In the first year, I worked closely with many of my fellow graduate students, but over the coming years, as we each got busy with our research, teaching, and other duties, we drifted apart and didn’t interact that extensively. Since I was one of only three graduate students in my broad area, and the only one in my year (and later I would be the only one in the whole program) I didn’t interact heavily with other people for mathematics. Interactions were mostly centered around social events where we talked about general stuff. I also wasn’t much into the small number of social/community events associated with graduate school life (like sports, or building events, or gaming).

Over the coming years, I did serve as an occasional consultant for students in other areas who occasionally needed knowledge or insights from group theory (my specialty) but this was sporadic and far less than the level of interaction between people who share an adviser and research area.

The upshot is that while I was friends with many graduate students I didn’t really consider it a community that I strongly identified with, nor did I find other meatspace communities or subcommunities that I identified with. I wasn’t unhappy about this but it did mean that there was a void of sorts in my life that could potentially be filled by other communities or sources of belonging.

Early interest in other intellectual domains (2007-2009)

I believe my initial interest in many intellectual areas outside of mathematics was spurred by a combination of factors. Even before joining graduate school, I bought books on many topics and read them regularly. After I joined graduate school, I would, in the winter, regularly visit the bookstore and sit there and read (and occasionally buy) books. This reading was haphazard, and I started out as fairly naive, taking a lot of the content of books I read at face value rather than pushing back at them or cross-checking their facts. At the time, for instance, I would read a book by Malcolm Gladwell (such as Blink) and accept everything he said. This wasn’t problematic because I wasn’t actually basing any decisions on what I read, but it still reflected a lack of intellectual breadth and skepticism. It’s hard to put exact dates, but my first bookstore-reading binge was in December 2007, and I read a lot of general-interest books in the summer of 2008.

I’m not exactly sure where and how I started identifying as a libertarian and getting interested in political philosophy, but I think one of the influences here was reading a book by Naomi Klein called The Shock Doctrine. After reading the book I Googled around for online critiques of the book, and found one by Johan Norberg for the Cato Institute. This led me to read more of Cato’s work and from there I discovered a lot of libertarian literature. As I started understanding more of what I believed and developing a clearer theory of the world, I was better able to integrate new things I read into that framework, rather than oscillating from one view to the next based on what I happened to be reading.

Personal evolution: continued interest in content creation, with increased focus on metrics

In May 2008, around the end of my first year, I moved from a shared hosting for Groupprops to a separate domain ( and set up the framework for the subject wikis. As I mentioned in my undergraduate experience page, I had worked a lot on content creation back in my undergraduate days, but had not been very systematic about it. As I worked on Groupprops and the other subject wikis, my thinking on the subject sharpened. I started writing up my thoughts and, starting January 2009, began posting them as sporadic blog posts on the Subject Wikis Blog. Two things were happening concurrently:

  1. I was trying to come up with a general theory of the reason the subject wikis were important, drawing largely on introspection and personal experience.
  2. Actual usage of the subject wikis was increasing, so whereas in 2008 Groupprops got about 20 pageviews a day, by April 2009 the number was up to over 100. At this point it was clear that there were other people who were referring to it, and it was no longer my personal fiefdom to do what I pleased with.

Both fed into each other: now that I had users, my theories were not mere abstractions that I could shape in a way that pleased me. Instead, they could be validated against actual user behavior. Going through the stats led me to become more convinced that some of the things that were personally important to me weren’t that important to my audience. This led me to restructure things in a way that I could continue meeting my own needs while delivering a better experience to the audience. The audience was still tiny and invisible. But it existed.

College Fellow and first year of teaching (2008-2010)

In my second year (2008-2009) I was College Fellow for the undergraduate algebra honors course. Starting with my third year, I had full teaching responsibility (you can see my teaching history on my teaching page). This was in some ways a sharp contrast to Groupprops and research: the audience that I intended to serve was right in front of me. In my first year of teaching, I didn’t have a clear theory of teaching. I was a pretty mainstream teacher, and even though I had some insights and thoughts about doing a better job teaching, I didn’t get around to implementing them.

April 2010- August 2011 (late third year and full fourth year)

I believe a number of things started clicking for me around the middle of 2010, though the ramifications of this clicking would take a long time to play out. In December 2009 I encountered the work of GiveWell, and through 2010 I made donations to GiveWell-recommended charity VillageReach. Around April 2010 is when I believe I first encountered EconLog and many other blogs in its dense network (Cafe Hayek, Marginal Revolution, Overcoming Bias, and Less Wrong). The summer of 2010 was a disastrous ACT Prep class teaching experience. Despite it going badly, I learned a few things from the experience that would shape my teaching experience for the years to come. Groupprops traffic continued to improve and I kept rethinking how to approach the creation of the site.

At this stage I had found many of the intellectual centers where the communities that I now interact with resided. But I didn’t know any of the people as individuals. I was largely consuming their content online and using it to feed my thought processes. Breaking into the relevant communities would be a different story.

The classes I taught starting 2010 (my fourth year of graduate school, and second year of full-time teaching) were qualitatively different from the classes I taught in my first year of teaching. In the Autumn Quarter, I implemented MCQ class quizzes, which, for good or bad, came to define my classes. At the end of the Winter Quarter, I implemented error-spotting exercises, another trademark of my teaching that would continue in the coming years. My student evaluations were overall more positive and I acquired more of a distinctive reputation (which may not always have been for the better).

The summer of 2011 (June 2011 to August 2011) I did not take up any additional teaching or mentoring duties, with the goal of focusing solely on research. That worked in some sense: I spent a lot of time adding content to Groupprops and exploring group theory questions of interest. But I was not able to focus substantively on a specific research question and make deep progress on that. So things were getting a bit worrisome on the research front.

There were a couple of other things that happened in 2011 that made me age a lot and feel more responsibility for what I was doing. Some embarrassing errors I found in Groupprops made me sit up straight and curse myself. At this points, hundreds of people were reading the website every day. Individual pages with glaring errors may have had over a thousand pageviews. This made me personally responsible for any misconceptions that arose from it. Groupprops was no longer a personal playground. What I wrote there, particularly in the important pages, had real consequences. So I went through all the important pages, carefully checking for errors. I fixed all I could find. I started maintaining an error log and error reporting mechanism.

At around the same time I discovered some errors in past materials and test questions I had given to students. There had been an error in an actual test I had given out to students in Winter 2011 (that I had discovered shortly after administering the test and had to adjust student scores accordingly). I discovered another error in a sample test I had given students for preparation historically. That these errors persisted despite my reviewing the tests made me realize the inadequacy of my review procedures. Again, in this case it did not matter, but I was not happy about my historical carelessness. In a sense, the carelessness was a product of my trying to do things that were somewhat more complicated than what I needed to do. But I needed to strive for more care. I felt more weight of responsibility.

After a trip home in August 2011 and a trip to Palo Alto to visit friends in September, I returned to the new academic year in 2011 with more general optimism but more confusion. I had made a speculative investment that, along with my various donations over the past year, had got my cash reserves down considerably. I wasn’t sure I either could or wanted to continue in academia. My increasing attraction to the ideas that would later be dubbed effective altruism led me to be increasingly skeptical of the value of doing so. Groupprops was getting to be a success of sorts, but I couldn’t make a living off of it. I didn’t have any original research results yet that could be used to get a thesis, and I was already entering my fifth year.

September 2011- August 2012 (fifth year): a turning point in many ways

The first two quarters of 2011-2012 went just fine. I had the Spring Quarter of 2012 off teaching. I embarked on two interesting projects around March 2012, both of which have had very different results. First, I collaborated with my former student Ruinan Liu on Vipul’s Classroom (YouTube channel and my site page about it), a collection of videos on calculus and group theory. The videos were decent and brought to the video mode some of the qualities I brought to the classroom, but they ultimately failed to make a splash, and after Ruinan left for graduate school in New York the project was abandoned. The project led me to spend a lot of time in ways that retrospectively seem inefficient — for instance, I spent a week figuring out various things about video and audio editing to re-edit the videos for noise removal, without any reason to believe that after all that effort, the videos would get pageviews. I don’t regret trying the videos but I believe that I should have given up on some unnecessary video-editing work.

The videos did get used later: I used them to supplement my teaching in Autumn 2012, and this led at least a few of my students to appreciate my teaching a lot more. I formed a lot more long-lasting relationships with a few students as a result of this. And given that the other things I was working on at the time weren’t that promising it may have been okay to spend time on video editing. But most likely I overshot there.

The second project I embarked on at the time was Open Borders: The Case (see my site page about the subject). Intellectually, it was a nice way for me to combine the moral/philosophical interest in ideas of liberty and efficiency with my interest in systematization that I had developed through extensive work on Groupprops. Despite the similarities with Grouprops and my desire to make the site a reference work, I chose WordPress, largely so it could look pretty from the get-go, and also because I wanted a blog section with potential collaborators. After I sent the link to Bryan Caplan and he blogged about it, the site started getting some traction. But more on that in next year’s story.

In the Spring of 2012, when, despite having no teaching duties, I wasn’t feeling passionate about research, but rather was spending time doing a mix of adding content to Groupprops, thinking through the Vipul’s Classroom videos, and working on Open Borders, I knew that staying in academia was not an option for me. The real, hard question was: should I finish my Ph.D. or walk away? I decided I will get through my Ph.D., somehow or the other. I narrowed the focus of my thesis, identified something I had already proved, and decided to write it up. The proofs turned out to be more complicated than I thought. My last 15 months (September 2012 to December 2013) were a struggle in motivating myself to get through a Ph.D. knowing that this was no longer my area of interest, knowing that the number of people who cared about the actual thing I would produce was approximately zero, and with a lot of uncertainty regarding where I would be headed next.

September 2012 – December 2013 (sixth and seventh year)

On the teaching front, I believe my last four quarters (Autumn 2012 to Autumn 2013) were my best, despite the fact that I was hugely distracted and stressed by the prospect of finishing up my Ph.D. and uncertainty about my future. I was able to integrate all the teaching tools I had developed over many years, ranging from MCQ class quizzes to error-spotting exercises to the videos I had shot with Ruinan, into the Autumn 2012 and Winter 2013 courses. The Autumn course was stressful because, despite all the material already being prepared, I had a large class size and so couldn’t get a lot done. And so I made very little progress on writing up my thesis. In the Winter, I made a lot of progress with writing up the thesis, but I also got to realize that it was a much more complicated undertaking than I had previously anticipated. While I had originally hoped to graduate in the Spring Quarter, it got pushed back to the Summer, and would eventually get pushed back to next Autumn.

Spring 2013 was an interesting time. I was teaching an entirely new course (linear algebra) and therefore had to prepare new materials (I had gotten a bit of a head start over winter break). The interesting thing was that, because of my increased maturity, I was able to develop much better teaching material despite doing it for the first time. The flip side of this was that I was overloaded with teaching material preparation and failed to make much progress on my thesis. I messed up one question on the final exam (luckily a student alerted me to the error early in the final and I fixed it). Otherwise it was a great experience.

My relationship with my thesis had reached a kind of “in it to finish it” mode. I had accepted my fate and was steadily working through to the end. I knew it would be an incredibly complicated object that I wouldn’t do in the way I would like to do it. The Summer of 2013 was a long slog. Winter 2013 I again had to teach linear algebra. At this point, the slog was almost enjoyable. The end was in sight. I knew I was close. I knew I had to work hard. I was okay with spending extra money on food so I didn’t have to cook and I could just focus on handling my thesis-writing and teaching. And I finished my thesis in time, and got my degree. And then started getting my mental space back.

Through this time, my alternative life was, slowly and surely, behind the scenes, taking shape. In late 2012, we got many more bloggers for Open Borders: The Case. I developed friendships and connections with people in the effective altruist, Less Wrong, and libertarian communities. My Facebook life was taking shape. Even if I didn’t spend a lot of time on those, they were becoming steady parts of my life. I still wasn’t sure what I would eventually do, but I had figured out that I’d work with Jonah on Cognito Mentoring and do contract work for MIRI on Optional Practical Training immediately after the Ph.D.

Addendum: I am grateful to a lot of people who supported me through graduate studies, despite the eventual divergence of my goals from theirs. Although I found myself a misfit for academia, I have been really lucky that all the academics at the University of Chicago have been very helpful. This starts with my thesis advisor, Professor George Glauberman, and includes professors Peter May (my second advisor), Jon Alperin (the group theory/representation theory expert), Shmuel Weinberger (department chair), Benson Farb (head of the graduate program at some point, when Peter May wasn’t). The graduate program administrator Laurie Wail, the undergraduate administrators Diane Herrmann and John Boller, and the office staff at the department were also very helpful and friendly. Ruinan, who shot the videos with me, some of my fellow graduate students, my parents, and some friends from my undergraduate days supported me at critical times in the process. I would probably have been a little worse for the wear for each person you’d remove from this list.

Basic information