This paper focused on students’ learning experiences in a flipped data science class integrated with peer instruction and just-in-time teaching. University students in Hong Kong participated in the research during the pandemic. Students’ perceptions of the flipped learning mode were investigated by a 5-point Likert scale questionnaire. According to the results, most students felt they enjoyed learning with the flipped mode since it allowed them to learn flexibly and independently. The course materials (i.e., instructional videos, lecture notes, and in-class exercises) were well-designed, and students perceived the materials were useful and user-friendly. As suggested by the students, more time should be given to the pre-class learning activities. Based on the suggestions, the researchers provide practical implications on improving teaching in the flipped science class. This study demonstrates how to flip a data science class and proves its value on students’ learning, providing implications for using flipped learning during the pandemic period. Copyright © 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG.