People at Civis come from a variety of backgrounds, and several of us left academia to work here. Myself, I earned a PhD in physics from UCLA, then spent four years doing cutting-edge cosmology research at the University of Chicago before coming to work with Civis. I still love physics, but coming to work at Civis was definitely the right decision for me. The data science research at Civis combines the best features of academia with the best features of industry.
When I started at Civis, I was surprised at how similar my day-to-day work experience felt to the academic work environment that I was used to. In both cases, you’re surrounded by really smart people working on the same (or related) projects. People are friendly and helpful. The most important thing about your work is the quality of the final product, not which specific hours you’re sitting at your desk. There’s even freedom to shape your own research agenda here — possibly more or less freedom than you would have in academia (much of one’s experience in academia depends on one’s faculty advisor / research group and/or grant status).
Of course, there are a lot of differences from academia as well. The output is certainly different. At the University of Chicago, the goal of my work was to learn and publish new and interesting information about the Universe. Here at Civis, we’re solving data science problems for all sorts of organizations, and the more practical aspect of the work here is a big plus for me. Another of the differences that I noticed was a different standard of openness. In academic research, I was used to being able to talk about nearly everything that my group was doing. In private industry, you can’t be quite as open, both because clients usually don’t want their business to be public knowledge, and because knowledge of what we’re doing and how we’re doing it is Civis’s biggest asset.
Another similarity between Civis and the academy is the meta-nature of the challenges. I’d say the biggest challenge we face here is the same that one encounters in any knowledge work: figuring out which question to answer. There’s a dozen different projects that I could be working on right now, but which is the most useful or most urgent? One of the great things about working at a small and growing company like Civis is that everyone here has a lot of opportunities to shape the future of the company, and picking which questions to work on is a big part of that.
Many of the skills and habits that I learned in academia transferred directly to my work at Civis. For example, I mentioned that the biggest challenge in this work is deciding which question to answer. You’ve also got to make sure that questions are framed in the right way. Is it clear what would count as an answer to the question? Would the answer be helpful for Civis, or for our clients?
Once you have the right question, it’s not enough just to find the answer to a question. You’ve got to test your answer to make sure it’s right. If your model works on one set of data, great! How about different data? Does it handle data that’s very different from what you used to create it?
Finally, once you have a solution, does it apply more broadly than the initial question? As much as possible, we build on past work. It’s good to be able to solve one problem, but better to be able to generalize that solution to future problems. As part of that process, we build up a store of software which is accessible to everyone at the company. All the code that we’ve written in the past is available to everyone who works here to help them solve problems in the future.
Civis also makes employee education a priority. We all learn a lot just doing our “ordinary” jobs, but beyond that, we put part of our time into learning new skills that we’re interested in. For example, I’m studying neural networks, because it’s not something I’ve had the chance to use before, but something that I think is really cool. We even have weekly journal club meetings where we discuss new research in statistics and machine learning, an event which will be familiar to anyone with an academic background.
I think one of my coworkers put it best when I asked him a similar question in my interview. He told me that working at Civis is what working in academia should be like, and I agree.