Data Scientist Henk Griffioen About His Work
Each data science role may be different, but our data scientists provide insight to help those interested in figuring out what a day in the life of a data scientist actually looks like.
Data scientists are responsible for discovering insights from massive amounts of structured and unstructured data to help shape or meet specific business needs and goals. Henk Griffioen, data scientist at GoDataDriven, gives valuable insight in his daily life as a lead consultant.
A Day in the Life of Henk Griffioen, Data Scientist
Right now, I’m working as the lead data scientist on an assignment for an international telecommunications company. We’re implementing data science and data engineering into the organization, and it’s a somewhat new topic for them, so we’re teaching them a lot.
With a project like this, the idea is to make our client sourselves self-sufficient eventually. We’re helping the organization to create their own data analytics team and showing them how to organize an agile workflow where they can build proof-of-concepts and bring those into production. Our goal is to ensure that they can eventually work independently.
Looking for use cases
Generally, we begin an assignment by exploring use cases with the business. We then build those use cases in six- to eight-week sprints that we then test. For instance, we’re currently testing a model that makes automated decisions about upselling and measure about money it brings in. With a proof-of-value like this, we can show the senior management how becoming a data-driven organization improves the company’s decision-making process.
Defining the actual added value is an essential part of the workflow, so we’re also implementing “assessment” as a specific part of every project. The world of data science has been hyped up in the past few years and data scientists like to do complex things. But what I often see is that what they build is not understood or used by the organization for which it was initially developed. That is why assessment is an important concern for me.
The complexity of this project is not so much in the models we build but in the organization of the workflow.
Learning new stuff
When it comes to my core science competencies, I develop those with the courses I teach and during our GDD Fridays.
Explaining a topic to a class really forces you to understand the material in detail. As I teach quite some courses (Data Science with Spark, Machine Learning, Deep Learning, etc.), there's a lot of knowledge to keep up-to-date!
During our monthly GDD Fridays we have a full day when we work on our own private projects. For instance, I have built a model to analyze and find relationships between emoji by scraping Twitter data. I am also working on a project that uses image-to-image translation, with which anybody with a webcam can “control” our CEO Rob. Unfortunately, the first version wasn’t very good, so I still have some tweaking to do, haha. Projects like these improve my technological skills.
The business side
I also want to keep improving on the business side of data science. To stay relevant as a discipline in two years, our data products should add measurable value to the business or result in new products available to end-users. Connecting to the right people and asking them the right questions is an important part of that process.
My goal is to help organizations develop their own data science competency.
Like many of my colleagues, I used to spend eveneings playing around with my personal data science projects. I love my field of work, but being behind a computer that much was making me very restless. Now, I feel much better and happier to jump on racing bike or play football. Sports clear my head.
My goal is to ensure that clients and their teams can eventually work independently.