• Home
• About me
• Publications
• Research
• Teaching
• Personal
• Academic CV
• Resume
• Blog

Hadas Kotek » Blog »

FAQ - My non-academic job market journey

Mar 2, 2024

The very first post I wrote on my blog was about my academic job market journey. I’ve written about various difficulties — funny ones and not-so-funny ones — on the academic job market. I’ve described my process of leaving academia in several posts in the past, but it’s still one of the things I get asked about most frequently in informational interviews. This post is structured as an FAQ and tries to address the most common questions I tend to get about my job search and prep process.

Did you always know you wanted a non-academic job?

No! I came into grad school and spent the first few years on the job market single-mindedly interested in academic jobs. I wasn’t thinking of other careers and wasn’t prepping for them. No one was suggesting them as viable options, either. I was told to “hang in there” and that “it’ll all work out”. I didn’t do any internships or have any relevant summer jobs. Any skills I had that were relevant for non-academic jobs were not acquired for that purpose; they just came about as the result of my academic interests and some accidents of who happened to be around and what they were doing that I just happened to be exposed to.

When did you decide to leave academia?

Leaving academia was a process—both the decisions, and the actions that needed to follow. The first time I said to myself that I wanted to leave was after a very unpleasant second job market cycle, described in the first section of this post. In fact, the spring and summer of that year were the worst time of my life. If not for the fact that my postdoc had a guaranteed second year, I’d have probably left right then and there. But I wasn’t ready to take action at that point.

I then had a great year at Yale, followed by a messed up year at NYU. The spring of 2018 (job cycle #5) was when I knew I was ready to leave, and, conveniently, a few months earlier I got my green card so my immigration status was no longer dependent on my job, and that helped a lot. But that spring and summer I had some health issues so it wasn’t the right time to change careers or lose my health insurance. So I took another one-year contact at Yale, with the very clear goal of achieving some kind of better career outcome — either a TT job or a non-academic job — by the end of that year.

How did you prepare to leave academia?

For several years leading up to my decision to leave academia, I was already applying very narrowly only to jobs that met a set of must-have criteria.¹ I tightened those up even more, and decided to only apply to a handful of academic job and to spend a minimal amount of time on prep.

In the fall, I reached out to a half dozen colleagues who had left academia before me and conducted informational interviews. Most of these folks were linguists who I had met at academic conferences over the years. A couple were from my PhD department. To this day I am incredibly grateful to them for taking the time to teach me all the basics! I really knew nothing and had no idea who would possibly want to hire me and what for. They told me what job titles to look for, what companies were hiring, what locations jobs were at, what the ballpark salary is I could expect. A few of them looked at my resume. One was particularly instrumental in my process—she commented on my resume more than once and helped me do interview prep. A few sent job ads my way that I would have never thought to apply for on my own, which helped me better define my search.

I spent some time in the fall and winter self-teaching using online courses. I studied basic SQL and basic Python, as well as intro to machine learning. In hindsight, I don’t really know if any of this made a difference. I also spent some time reading about technical interviews online, but that was definitely not the right level to worry about for the jobs I was going after. (The technical interviews I did were overall pretty basic.)

In early spring, I made a resume using the same Latex template as my CV. I had multiple people look at it. Fortunately, these were friends and colleagues who were able to point out things I had done that were not on my resume but should have been (highlighting the importance of getting people who know you to look at your resume; a review from a third party can only get you so far if they don’t know you). I spent a bit of time collecting and uploading some materials to github and adding links to my resume — I had some analysis scripts in R, some documents I learned to call design documents and annotation guidelines (though as an academic I didn’t use those terms, they were just handouts when I made them), and I had a few experiment designs and results I could show.

I did a bit of interview prep, mostly to be sure I was able to speak about everything on my resume and to prep for behavioral interviews (which are definitely not something I could just do without prep, it’s a very specific genre). I did this both in the form of writing detailed notes for myself and in the form of practicing with friends. The upside of having been on the academic job market for so long is that I was experienced in interviewing in general, so I didn’t feel as overwhelmed as I remember feeling when I first went on the academic job market. I also read about technical interviews but didn’t really find descriptions that seemed relevant for the jobs I was applying for so I wasn’t sure what to expect before I started applying.

I also created and populated my LinkedIn profile, and I cleaned up and locked my social media accounts.

How did you find jobs to apply for?

I had some links to job ads sent to me by friends. I went on job boards for the major companies I had learned in informational interviews might hire linguists, such as Google and Amazon. I set alerts on keywords like linguist and language. I searched relevant keywords on LinkedIn. Whenever possible I tried to have internal referrals for jobs, rather than apply cold, but I did a bit of that, too.²

What was your actual process like?

I had a pretty unusual process in terms length and outcomes, which always makes me extra wary of offering specific advice in this domain. I was given very good advice (which I didn’t take) to apply to some jobs that seemed potentially interesting, even if not perfect, to gain some experience applying for such jobs.

Instead, roughly a couple of weeks after I had heard that I wasn’t hired off the only academic shortlist I made that I was really interested in (in early March), I started out by applying to exactly one job (through a referral), which checked all the boxes for what I was looking for. Within a month I had phone screens with the recruiter and hiring manager, a couple of phone interviews, and an on-site interview. I didn’t get that job.³ I then broadened my search a bit and in early/mid April I applied to four jobs.⁴ Within a month I had three offers (titles ranged from ontologist to computational linguist (Hebrew) to natural language annotation project lead; compensation, location, and conditions varied greatly).

I got my offer from Apple the day after my on-site interview, while I was doing an on-site with another major competitor.⁵ I did not negotiate my compensation (generally a mistake, but in the context of my personal history it made sense; and it was already a very good offer, as it turned out); I proposed a start date roughly 2 months out, though in hindsight I probably could have suggested a date farther out and that might have been easier. Regardless, I moved out to the West Coast in mid-June and started the job on July 1.

In all, that’s about 2 months from first application to signed offer, and ~3.5 months to actually starting that new job. It was fast, smooth, and successful — everything my experience on the academic job market was not.

I was done with teaching and grading at Yale around mid-May, which didn’t leave a lot of time to unwind or wrap my head around all the changes that were happening in my life, but on the other hand once I was settled into my new apartment and unpacked enough,⁶ I was finally able to adopt cats—which I had been holding off on for years while I felt like my employment and living situation weren’t stable enough, so at least that was great.

What were the interviews like?

The informal chats with the recruiters and hiring managers were always pleasant and straightforward, they just asked basic questions about my resume, and told me some information about the job I was applying for, such as the day-to-day and what the team composition was.

The one thing that almost everyone was most interested in was the project that led to the publication of Gender bias and stereotypes in linguistic example sentences (Kotek, Dockum, Babinski, and Geissler; Language, 2021). That project involved the creation of a corpus of ~30k example sentences from 1.1k articles published over 20 years in three leading Linguistics journals (corresponding to all numbered examples from all articles published over that time period). From that corpus, we extracted ~25k 3rd person human arguments.⁷ We got a small grant to hire 25 Yale undergraduate students to annotate each argument along several dimensions of interest.⁸ The result was a large-scale analysis of aggregates as well as trends over time, demonstrating gender bias along multiple dimensions in linguistic example sentences.

To be clear, interviewers were not too interested in the results themselves (beyond a basic description) but in the process: how did you choose the journals? how did you choose what dimensions to look at? how did you recruit and train 25 students to produce consistent annotations? how did you deal with conflicts or discrepancies in the data? what was your quality control process like? what were the logistics behind taking the corpus and assigning each data point to someone to annotate? how did you collect and review the results? how did you handle questions and ambiguities? how did you choose what analysis methods to use? In this context, I was able to describe an onboarding and training process and annotation guidelines for the Yale undergrads, and that was directly applicable experience for most jobs I was applying for.

Aside from that, job interviews generally tended to be divided into:

• linguistic knowledge interviews, which were never about linguistics or theory but rather about broad thinking about language data and phenomena,
• knowledge representation interviews, which were about ontology/taxonomy/knowledge engineering and occasionally also annotation project design and approach to guidelines,
• project management interviews, which were about, well, project management experience,
• product sense interviews, which were about how to approach products from a business-minded perspective,
• technical interviews, which tended to consist of one or two python problems and perhaps writing a simple SQL query, or data analysis questions, done in real time on the board, and
• behavioral interviews, which are those “tell me about a time when…” story-based questions, which you should definitely prep for because most of us can’t just tell STAR-based stories for major events in our lives off the top of our heads.

The question of why I was leaving academia almost never came up, and when it did a simple answer was enough. Something like “there are things I really enjoy about academia but there are things I want to prioritize now that I haven’t found there, such as the ability to choose where I live, better compensation, fast-paced team-based work, and a high impact product”.

Interviews always ended with “do you have any questions for us” so it’s good to come prepared with a few good ones. Depending on who you’re speaking with, you could ask about team culture, growth opportunities, day-to-day in the job, what would make for a successful first year in the role, or why the role is available in the first place. I always like to ask “is there anything I haven’t told you that would help you make a decision” and sometimes also “what else should I ask”. Those two questions tend to get the best reactions and to elicit the most useful information.

What part of your academic training was most relevant to jobs and interviews you had?

From my grad school days, the most relevant thing was the experimental experience I had: operationalizing a research question, experiment design, writing guidelines, recruiting participants, data visualization, data analysis, thinking about power, how to identify cheating behavior or other trends or issues, how to choose which tools to use.

Also: presentation and writing skills; being able to identify interesting problems and to justify why they matter and why others should invest in them; self-teaching; being able to identify resources independently; persisting in the face of uncertainty; being able to keep track of multiple parallel projects, including figuring out timelines and project needs.

From my time as faculty (mostly): leading and directing teams (of RAs and TAs); being able to teach both basic and advanced materials, knowing what level the audience needs to be addressed at, being able to give examples and point people to resources; mentoring; ability to design and conduct semi-structured interviews (fieldwork); grant writing and budget administration.

How long would you recommend looking for a job?

The advice I was given was to start looking 6-8 months before you want to start the new job, and although my process was much faster, that still seems generally reasonable to me. The details will really depend. If you’re constrained in location or type of job or compensation you’d entertain, or if the market is saturated in your area because of layoffs, or if there are widespread hiring freezes, or if you need immigration support, then you might need longer. The more criteria you are willing or able to compromise on, the more options will be open to you. But there will always be factors outside your control, so it’s hard to know ahead of time.

If you’re looking for your first job, it’s also harder because you need to justify a career transition and you need to find someone who’s willing to take a chance on someone without direct experience in the field. That again might take longer. In this context, it’s worth remembering that your first job doesn’t have to be your dream job. It’s much easier to get another job once you are already employed in the field. It’s totally fine and normal to leave a job after 12-18 months for a better opportunity, so you’re not stuck just because your first job isn’t the perfect fit. In general, your career path is much more flexible outside academia and there will always be multiple options for your next career move at any point.

I never got to this point; but if that narrow set of applications I had started with hadn’t yielded the right results, the two next steps would have been (a) re-inspecting my materials to try to figure out whether there’s something about them that wasn’t landing right with hiring managers or recruiters (in case I wasn’t getting interviews) or if I needed more interview prep (if I was getting interviews but no offers), and also (b) expanding the scope of what I was looking for in terms of type of job and location.⁹

Did you share with your advisors that you were actively looking to leave academia?

I shared my experiences with my colleagues at Yale in real time. They were nothing but supportive. Close friends and family and some of my non-local colleagues knew, too. I wasn’t keeping it a secret or asking people I talked to not to repeat the information, but I wasn’t posting it on social media or anything, either. I’m pretty sure I didn’t used the “open to work” label on LinkedIn, but not for secrecy reasons. As I mentioned, it just all happened very quickly and I didn’t end up needing to do some things that would have made sense if it had taken longer.

I had the opportunity to visit MIT and see my grad advisors in person shortly after accepting the offer, so I just basically announced it to them and to the world (on social media) as a done deal at that point. I think I took many people by surprise; that said, it wasn’t a secret to my broader community that I had had multiple bad experiences in previous years, so it shouldn’t have come as too much of a shock.¹⁰

Something about your interactions with people changes once they know you’re leaving. I think that’s a fair concern to keep in mind. Sometimes you have good reasons to suspect people will react poorly and sometimes you just don’t know. You can absolutely pick and choose who to share important life decisions with, and when.

So, what do you know about immigration and non-academic jobs?

First off: none of what follows should be construed as immigration advice. Always verify with other, reliable sources for anything as important as your immigration status!

I don’t know much, to be honest. I was under the impression that it would be much harder to get an H1-B visa and a more stable immigration situation in a non-academic position than in an academic one. I never did much to confirm or disconfirm this belief. I did know that if you had an academic job, then the visa would be more or less guaranteed and there are no quotas, whereas there are quotas if I went for a job in tech.¹¹

During my second year at McGill, I decided to initiate a process to self-sponsor a green card, through the EB2-NIW category. On the heels of that very unpleasant experience on the job market I alluded to above, I figured I shouldn’t leave my immigration status up to the whims of random uncaring academics. I hired a law firm that specialized in green card applications. The whole process took close to two years, end to end, most of which was spent either in the ‘gathering documents’ stage or the ‘waiting to be processed’ stage. When I got my Yale job, the main thing I negotiated for was the H1-B (rather than J-1) visa, not salary or other conditions. At NYU that was more straightforward and didn’t require much negotiating. I got my green card during that year.

In short, I only applied for non-academic jobs after I had a green card, and never even attempted to find out more about my options before then.

By now, I’ve come to learn that large companies tend to have their own in-house lawyers, or they have immigration firms on retainer, to handle employee immigration issues.¹² They will often hire employees without regard to immigration status and have the lawyers figure it out. I even know of cases where an employee was temporarily located in an office in another country than intended until their immigration status was figured out. I also know that smaller firms are less likely to be set up to hire people who require visas, or to know how to handle H1-B sponsorship, or what OPT means. It doesn’t mean it’s not possible, but in general bigger firms are a safer bet, it seems. But — as I said at the top — I’m no immigration expert, so take all of this with a grain of salt, and ask your own questions.

 

Notes