• Transport mode algorithm improvements

Our current transport algorithm detects when an individual is 'in vehicle', travelling 'by bike' or 'on foot'. To do so, we evaluate millions of coordinates on a daily basis. Coming from a pure feature engineering model, the current challenge is to improve the algorithm. You can put your knowledge about hidden markov chains or neural networks to the test. On top of that, we have access to the realtime coordinates of De Lijn (bus transport) to help improve detection of bus trips'. By mapping coordinates on railway roads, defining the 'in train' transport mode can be defined.

‍

• Proactive recommendations

Today, multiple metrics about customer loyalty, competitors visited by a brand's customers or popular hours (to name but a few) are calculated on a continuous basis. Too much information to see the forest for the trees... In this role, you will explore how Large Language Models (LLMs) can be leveraged to generate proactive recommendations on which metrics deserve deeper analysis. Your research will focus on transforming insights into actionable steps, enabling smarter decision-making and driving business impact.

What’s on our wish list?

• you have a bachelor’s or master’s degree in computer science or related field
• fluent in python
• fluent in developing with databases (MySQL)
• knowledge of Git and Linux
• you speak fluent Dutch
• hands-on attitude and comfortable in a fast-moving start-up environment
• experience with data collection and processing is a plus

Experience with any of the following is considered a plus

• experience with (Google) Cloud technology
• geo-spatial data processing
• data engineering technologies (Dataflow, Bigquery…)