Yesterday I saw Mitacs tweet about the Beyond The Professoriate conference, and I retweeted it to spread the word (I spoke at one of the first editions of Beyond The Professoriate and think it is a great initiative), and added a little thought about how it was a collaboration for me that launched my career in tech.
Hindsight is 20/20! Can you share what sort of collaboration it was, and how did it transition you to tech?
— Mitacs (@MitacsCanada) March 28, 2018
This post is a reflection on the collaboration I mentioned, and also on a bigger problem: we often fail to recognize which skills and experiences are transferable to non academic work. When speaking to PhD grads looking for jobs outside academia, I always stress how important it is to translate your experience to fit a job description, and communicate in ways that make it easier for companies to see you fitting into their organization. However, I also failed to do exactly this with the most applicable and transferable experience I could have asked for during my PhD.
Not surprisingly, this is the experience that ‘got away’, and never made it onto my CV (still hasn’t!).
The collaboration that sparked my career in tech
For my dissertation, I studied the effects of low dose embryonic alcohol exposure to social behaviour development, and brain chemistry. A real mouthful, in essence attempting to model and study effect of consuming just a little bit of alcohol during pregnancy.
We were using zebrafish as a model organism, and because this related to social behaviour I was pioneering work on recording behaviour in groups of fish, together in one tank. Social behaviour between fish separated by a barrier had already been well documented, but I felt this was still quite artificial and it was worth exploring a more natural setup where they could interact freely as a group. This required a way to record location data for 5-15 fish at a time, in the same tank. There were already some commercial software applications that could track location, but only for one subject at a time. We were using Ethovision, by Noldus, to track single fish. We had developed a good relationship with them, and I brought our challenge to their attention. At the time, adding the necessary features wasn’t going to happen quick enough to make it useful for my research (unfortunately), if I wanted to graduate within a reasonable amount of time. Their software at the time was brilliant for one subject, I just needed more.
A fellow PhD student had come up with a simple application to solve our problem. With it, we could manually identify each fish in each frame. With 30 frames per second, it goes without saying that this was too time consuming to do in any kind of high resolution. Even recording the data every 30 seconds was time consuming enough to take months to complete a study. What we really needed was a similar application where the subjects could be tracked automatically, but none of us had the skillset to develop it ourselves, or time to learn to code with the risk of still not succeeding.
Some time later, by complete chance, I met James McCrae thanks to another research colleague in the lab. One conversation lead to the next, then to an idea, which eventually lead to a prototype. Before long, we had a beta version of an automatic tracking system for our fish.
My first product – The REAL Fish Tracker
The first application developed by one of our lab colleagues was called “Fish Tracker”. This was the application that required a lot of manual work to generate the data. As a friendly joke, the new application that came after it was baptised “TheRealFishTracker” by its developer. It was an exciting project that helped my research take leaps forward. But what was even more exciting was the experience of seeing this application go from concept to a real thing, something I and fellow colleagues, could use. At the time, I had no clue that I was playing the role of user, QA, and product manager all in one throughout its development and iterations.
Together with my collaborator (a.k.a. my dev team of 1!), I discussed the user needs, features required for a minimal viable product (MVP), specs, acceptance criteria, and so much more. We even took an agile approach to the development. However, we didn’t have a Jira board, and we didn’t have an official roadmap, and only in retrospect do I know to attribute these terms to what was a string of conversations and emails back then.
Regardless of whether or not we used the “right” terminology, we got the job done and I from my perspective all credit goes to the developer. What we ended up with was an application that tracked a variable number of fish, and provided me with a .txt output for the data in the form of coordinates. That then needed to be, processed, analyzed and summarized in R, but that is a topic for another blog post.
I would never have thought to include my product management experience from my PhD on my CV to apply for jobs in tech. At the time it wouldn’t even have crossed my mind, but now that I look back it makes no sense that I wouldn’t have. There were two reasons for this not to make it on to my CV: Mostly, because I never spotted that these experiences were relevant and valuable. Secondly, I would never have thought a job like this existed.
It was as solid an experience in product management as I could have gotten in any company. The problem was, I had no idea I was operating as a product manager.
Product management has existed in various forms from the 1930s, and has founds its way into tech since at least the 1990s. I defended my PhD in 2012, so in theory I could have applied for a PM role with some very transferable and applicable skills from my PhD. I just didn’t know the job existed. It’s very OK that things worked out differently, but what is worth noting now is that I had relevant experience managing the development of a software application from start to finish. I completed several rounds of QA, I even did user testing (mostly on myself, but still), and I discovered and reported bugs, wrote specs for new features, and learned all about breaking changes.
The real challenge of transitioning to a career outside academia is not that during a PhD you don’t acquire the right skills. Rather, it is more likely that you don’t know how to translate relevant experience to a non-academic CV. Beyond that, the other challenge is knowing the types of careers available that match your unique set of skills. Not everyone comes out of a neuroscience PhD having collaborated with a computer scientist, but most of us have solid project management experience, analytics and data management experience, and an arsenal of communications skills.
Can we really say that this chance encounter that accelerated my research also sparked my career afterwards? This all happened a year or so before I joined a software startup. I still credit my job at Papers to being the first step in my career in tech, leading me to product management. Unofficially, I think it all started before then. We are ultimately drawn to things we love. I’ve always been drawn to problem solving, data, and making workflows better. Wrapping that altogether, I’ve worked on projects and products that involve all of these aspects so far.
image at the top by John Schnobrich on Unsplash