Since the widespread emergence of generative AI, arguments about the right approach to teach introductory programming are often presented as one of two extremes: “vibe coding” or prompts first [1] or “trad coding” first.
This project will explore the space between these two extremes by implementing one or more proof-of-concept programming platforms for novices that mix these two paradigms.
For example, a learner might provide a natural language prompt to create a program but pedagogically important parts of the program might be left as “blanks” for the learner to complete themselves. Alternatively, a natural language prompt provided by the learner might lead to a Parson’s Problem [2] to be completed by the learner.
Projects on this topic could be tailored to either Year 4 or Masters level students.
[1] B. N. Reeves
et al., ‘Prompts First, Precision Later: Reviving the Vision of Natural Language Programming for Computing Education’, in
Proceedings of the 25th Koli Calling International Conference on Computing Education Research, in Koli Calling ’25. New York, NY, USA: Association for Computing Machinery, Nov. 2025, pp. 1–8. doi:
10.1145/3769994.3770039.
[2] B. J. Ericson
et al., ‘Parsons Problems and Beyond: Systematic Literature Review and Empirical Study Designs’, in
Proceedings of the 2022 Working Group Reports on Innovation and Technology in Computer Science Education, in ITiCSE-WGR ’22. New York, NY, USA: Association for Computing Machinery, Dec. 2022, pp. 191–234. doi:
10.1145/3571785.3574127.
[3] S. Thorgeirsson, T. Ewen, and Z. Su, ‘What Can Computer Science Educators Learn From the Failures of Top-Down Pedagogy?’, in
Proceedings of the 56th ACM Technical Symposium on Computer Science Education V. 1, in SIGCSETS 2025. New York, NY, USA: Association for Computing Machinery, Feb. 2025, pp. 1127–1133. doi:
10.1145/3641554.3701873.
