A week in the life of a data scientist at Planning Services – decoded

Discover what a week in the life of Rijk van der Meulen, data scientist at EyeOn Planning Services, looks like!

 

14 commits to our codebase and 4 pull request reviews

Within EyeOn we develop our own forecasting and inventory plugins. At the time of writing, we have developed over 30 plugins, containing tens of thousands of lines of Python code. Each Monday we have a so called “sprint” meeting where we distribute tickets among the developers. A ticket is essentially an item that outlines a problem or a new request. This week I got to work on adding Spark support for one of our forecasting plugins. This feature allows us to process data faster by splitting the work up into chunks and assigning those chunks across computational resources.

While I was working on this feature, other colleagues haven’t been sitting on their hands either. So; time for code review! A code review helps us to ensure everyone is aware of code changes. Also it allows us to check the code quality and discuss potential modifications. Apart from this process being essential for the reliability of our plugins, it’s also a great way to learn from other developers.

 

1 new Honeycomb project created to forecast the demand for over 75,000 DFUs

All our data science projects are executed on our “Honeycomb” platform. It is built on Dataiku, an industry standard data science platform. Honeycomb simplifies collaboration and enables us to run data science projects quickly and effectively. Recently we started a new phase in a project to forecast the demand for a multinational in the consumer products industry. The challenge here is to predict the demand for over 75,000 demand forecasting units (DFUs) using both historical sales data as well as promotional features. In the weeks to come, Dirk and I will be working on finetuning our machine learning algorithm to deliver the best possible forecast.

 

3 customer meetings

With our Planning Services team we provide recurring outsourced services focused on forecasting, inventory optimization and actionable end-to-end supply chain insights. A fun part of running these services is that it allows me to have a lot of interaction with the customer. At present Bas and I are running a service for a global Japanese pharmaceutical company focused on reporting and robotic process automation (RPA). On a daily basis we work closely together with their supply chain team on lots of interesting topics.

 

4 table football games, 1 evening out with colleagues, and an uncountable amount of fun

I once summarized the culture at EyeOn as an ambitious group of people who really want to achieve something while having a lot of fun together. There is absolutely no place where this culture is better amplified than in our table soccer room. Yes, all of our consultants go to great lengths to deliver the best possible results for our customers but even this dedication falls short compared to the commitment to not lose a game of table soccer. You can thus imagine my pain confessing that I lost 3 times to Martijn last week… All jokes aside, I count myself lucky to work in an office with such caring and fun people. For me this is one of the key contributors to finding a deep sense of joy and meaning in my everyday work.

Hope you enjoyed reading this post! If you have any questions or comments, don’t hesitate to reach out.

 

Rijk van der Meulen, data scientist at EyeOn Planning Services

Search for