Dr. Martha Austen graduated from the department in 2020, and shortly after found a job in industry. She agreed to share about what she's been up to since graduation and give some tips about the job search process outside of academia.
Tell us about your job - where are you working and what's your title? I'm working at a hedge fund called Weiss Asset Management as a "Quantitative Research Analyst".
What was the job search process like? The job search process took longer than I expected. I was interested in becoming a data scientist, so I mostly applied to things with titles like "Data Scientist" or "Data Analyst", and I wasn't picky about what domain these jobs were in--I wasn't especially interested in making sure my job had anything to do with linguistics, and in fact was really excited to get to learn about a new domain.
My specific process was that every day, I searched for jobs on Indeed with keywords like "data scientist", "data analyst", or "researcher", and then I made myself apply for a certain number of jobs each day. (I think around 2-5.) It's a lot quicker to apply for industry jobs than academic jobs--you do want to tweak your resume/cover letter slightly for each job you apply to but this usually only takes about 15 minutes, and gets faster the more jobs you apply for. I also got advice from OSU's Career Services department--they were really useful and did some very detailed editing of my resume and cover letter for me.
I spent about 5 months searching (the first 2 months very casually) before I got any job offers, but when I finally did get an offer I got three at once, so that was nice.
What do you do on a daily basis as a quantitative research analyst? I work on trading energy-related commodities (e.g. oil, natural gas, electricity). (Side note: I don't do any trades that make me feel like I am perpetuating our reliance on fossil fuels (besides partaking in capitalism generally), i.e. I'm not doing anything like funding oil wells.) This basically involves trying to predict how much different forms on energy will cost at different points in time and then making bets based on those predictions. These costs are based on supply and demand, so you can try to predict things that will affect the supply of a commodity (e.g. for electricity, if it's cloudy you can't generate as much electricity from solar panels) and the demand (e.g. for natural gas, if it's cold lots of people will turn on their heaters, which will use a lot of natural gas). We usually take a statistical approach to this, so I might collect a bunch of data about the price of energy commodity X and then try to see how this correlates with Y or Z.
Most of my day is spent doing data analysis in R, making documents about my findings, and going to meetings (there are many, many more meetings than in grad school). The data analysis is not radically different from the kind of analysis I did in grad school, except that it is about energy instead of about linguistics.
What do you most enjoy about your job? What do you most miss about academia? I really like getting to spend most of my day doing data analysis, which is why I wanted to be a data scientist/analyst in the first place. (Looking at data and making plots was definitely one of my favorite activities in grad school.) As compared to academia, I really like that the turnaround time on things is much faster--a project usually takes no more than a month, and there's nothing like having to wait six months to get reviews back from a journal. I also like the lack of bureaucracy: I never have to write any IRB protocols, and if I want a book or some supplies relevant to my job the company will just buy them for me instead of making me submit a grant proposal. It's also very nice that they pay me a lot more than what I would have made as a postdoc. And I enjoy getting to learn about energy.
Some things I miss about academia are discussion groups, conferences, and doing science. It is a lot more motivating to have your research question be something about how the world works instead of just "how can we make more money?". This a bit opposite from most people, but I also miss the working hours in academia: in grad school, I usually only worked Monday-Friday from 9-5, and now I'm working more like 50 hours a week.
Of the skills that you acquired in grad school, which have been the most important and helpful for your job? Definitely R and skills related to data analysis, like statistics and how to communicate results effectively.
Do you have any advice for students who are considering working in industry? Industry is pretty nice! If you are considering industry, I would recommend periodically checking for jobs on Indeed in the city of your choice just to get an idea of what's available/how frequently new jobs are posted/what kinds of skills are most commonly requested. (Another nice thing about industry is you can choose where you want to live! And you can choose to change jobs if you don't like the company you're currently working at while still getting to remain in the same city.)
Here is some advice for people who are applying to data scientist/data analyst jobs specifically (Note that I think most linguists who do statistics/use R should be able to get a data scientist/analyst job -- some of these positions want you to be able to do fancy machine learning stuff, but there are a lot that basically just want you to be able to make a plot and do a regression):
- Learn some SQL so you can put it on your resume. (I did an online course.) It is pretty easy to learn the basics (definitely easier than learning R or Python) and is in high demand.
- A lot of job ads say that a degree in computer science is a requirement. Ignore this. Some places really mean this, but a lot of them don't, and you don't know which are which.
And here is some generic job search advice:
- Usually the people reading your resume work in HR and don't know anything about your specific job role, so make sure the phrasing on your resume matches the job ad exactly. For example, if a job ad mentions "linear regression", make sure your resume says "linear regression" and not "linear models" because the HR person won't know that they're the same thing.
- The more you do job interviews, the better you will get at them. The first few ones you do will probably be awful; think of them as practice.
- Don't get too attached to any one job you apply to; the chances of you getting any particular job are slim. (This is why you should apply to lots of them!) This advice is also true for academic jobs.
- A lot of job advice talks about the importance of networking, but it is also very possible to get a job without having any connections. All of the job offers I got were from job postings I just found on Indeed and didn't know anyone working there.