We explore the integration of artificial intelligence with digital game-based learning in the context of teaching the database query language SQL at a business school. We extended the game SQL Scrolls with an AI based personalised recommendation algorithm that provides task recommendations using a model which considers player performance and task difficulty. Our evaluation with 41 participants of various backgrounds highlighted the possible impact of the class environment, previous programming experience, group composition and size, and session length on engagement and playing speed. The students needed 42 to 62 s per SQL task on average, which is a fast pace. High levels of interest and engagement were evident, with most participants giving positive feedback. This personalisation led to good playing outcomes, with students progressing fluently through the game.