Marloes van Eijk
What is it that you do exactly?
I work in a relatively new department that uses data-driven reports to track down criminal and inexplicable assets. Our objectives are to trace criminal and inexplicable assets, combat tax evasion, recover money owed to the state and provide the government with tools for meaningful intervention. We deliver three standard products: network analyses (these allow you to see what a suspect's network looks like); insight into a suspect's assets and sources of income (is there anything tangible, like property or cars, that can be confiscated); and a product that provides greater strategic insight into the money-laundering practices within a specific region. These products in turn enable our collaborative partners – including the police and the Dutch Tax & Customs Administration – to do their jobs more effectively. Only those with special powers and when the individual or business in question is officially under suspicion can request such reports. The function and task of the applicant affects the amount and variety of information that may be displayed about a person. An inspector or collector of the Dutch Tax & Customs Administration may, within an ongoing investigation, make a request for a report. However, no information from police files can be included in this report. For requests from the police, where all available sources are consulted, a claim by the Public Prosecutor is always required.
I work on the R&D side, which means my focus is on the improvement and innovation of our products. Currently, for instance, I'm working on a project involving text analysis. I'm trying to develop a kind of search engine that can find and retrieve relevant documents from an enormous digital archive, which will give us a more complete picture of the suspect. I let the data do the talking. We try to use the data to determine the scope and direction of a given investigation, so that detection capacity can be deployed as effectively as possible.
What does a day at work look like for you?
Every day is different, which is what I like about my job. I sometimes spend entire days just working at the computer, but on other days, I'm out and about talking to our partners to gain knowledge and pool our strengths, so that we're not all off in our separate corners trying to reinvent the wheel. For instance, I organise brainstorming sessions to help identify needs and ideas, which we then try to put into action. I also organise sessions to address specific themes in our field; these help us familiarise our colleagues in other parts of the organisation with the possibilities offered by data science.
Who are your counterparts/stakeholders?
We cooperate closely with the National Police, the Tax & Customs Administration, the Fiscal Information and Investigation Service (FIOD), the Financial Intelligence Unit, Special Investigative Services and the Public Prosecution Service, among others. These partners actively assist us in developing our data-driven products; they're also the ones who use our products.
What’s your favourite aspect of the job?
I like that it allows me to be involved in substantive and practical aspects, and that it combines lots of consultation and visiting counterparts with sitting at the computer developing models by myself. I wouldn't want to spend all my time looking at a computer screen. I also like that it's a relatively small department that includes delegates from all the different collaborative partners. The lines of communication are short and that works really well. I'm not a subject-matter expert; I know a thing or two about data, but not that much about criminal networks or financial crimes. This way, when I run into an issue, I can just walk down the hall to get the information I need. Also, I also feel it's important to contribute to society in some way.
Where do you see yourself in 5 years?
I don't have an exact plan mapped out. I've only had this job since January. However, I do want to keep developing my talents, so I'll see how it goes and what I might need in future. There are plenty of challenges here – even just keeping up with all the developments in my field.
Why did you start the master's in Artificial Intelligence at Maastricht University?
After earning my bachelor's degree in Artificial Intelligence in Nijmegen, I decided to come to Maastricht because of its Problem-Based Learning (PBL) system. PBL has closer ties to professional practice and takes an interdisciplinary approach to problem solving, which I found tremendously appealing. I like that the courses cover a mix of disciplines: psychology, IT and even some philosophy. Another advantage to Maastricht was that I could attend courses from the master's programme in Operational Research as well.
What advice would you give current students and recent alumni of the master's in Artificial Intelligence?
There are loads of jobs for data scientists out there, as I discovered after graduating. You'll likely be fending off constant enquiries via LinkedIn, but take the time to shop around. You have the luxury of really being able to pursue your heart's desire.
If you had another chance, what would you do differently as a student?
Not that much. It was a shame I didn't live in Maastricht while studying there, but that was because I had decided to spend a few months travelling between my bachelor's and master's programmes, so I didn't have a room in Maastricht. Also, I chose to do a work placement in Amsterdam, which I'm still quite glad to have done. So no regrets, really.