Since the digitalisation of urban mobility, such systems generate large and heterogenous data sets. Handling these however is still a major challenge. In order to help people making sense of urban data, skills in data literacy are needed. In this paper, we present results and insights from a set of mobility visualization projects students from computer science and design developed over one semester. The resulting prototypes visualize a) user generated bike, b) population movement, and c) public transit passengers. After a brief justification on why we find urban mobility a practical domain for teaching data literacy, we introduce the course and describe its set and setting. For each project, we report on the challenge, the data, the visualization prototype, and the impact these projects had. Our contribution is two-fold: In the description of the problem-oriented course and its results, and in the reflection of our approach and our lessons learnt.