Q: What do you do at Jumio and what does your role involve?
I’m a data scientist in the risk team. We develop machine learning (ML) models for estimating the risk of fraudulent documents.
Q: What are the most interesting / fun bits of the role?
As someone with a pure ML background, I’m always thrilled to learn about the application domain. At Jumio, I’m constantly learning about a variety of ID documents coming from different countries. And I find it extremely interesting to work on the challenge of detecting fraudulent documents.
Q: What do most people get wrong / misunderstand / not appreciate about the role?
Sometimes people think that ML or AI do the work for you, kind of like magical black box solutions. Well, that’s very far from reality. One needs to know their data and understand what the problem is and what data and why you want to use a certain model and not others, otherwise you have garbage in, garbage out.
Q: What’s the secret to being good at your job?
I would say continuous learning. You should always be up-to-date on what’s new and happening in your field to be able to improve things you’re doing.
Q: What didn’t you expect about the role / what surprised you most?
The variety of possible fraud cases (the many ways of how one can commit fraud) and consequently the variety of things we can implement to stop it.
Q: What challenges did you face and what did you learn from that?
It is the first time I have had a fully remote job. And since you do not regularly meet your colleagues at the office, there is the caveat that you might start to feel isolated. I’ve learned that you should keep active and do not hesitate to ping people and just talk. And it helps also with getting familiar much faster with how things work at Jumio.
Q: What would you say to potential employees wanting to work at Jumio? What qualities should they have?
Apart from the professional qualities required for the role, they should be open-minded and eager to learn.
Q: Can you tell us something not a lot of people know about you or the job you do?
It was not at all my plan to do a Ph.D. in ML. I wanted to do the additional master’s to learn more about big data and go back to working in that industry. But then I had a couple of ML courses in the program and this is how my ML journey with a Ph.D. and two postdocs on the topic started.