Iterative Design of a Teaching Assistant Training Program in Computer Science Using the Agile MethodGlobal
Facing soaring enrollment and disruptive educational technologies, computing education increasingly relies on the contributions of teaching assistants (TAs), hence the critical importance of high quality TA training. However, the design and implementation of TA training in computer science face substantial barriers, such as the lack of experienced TA trainers and the scarcity of relevant training materials.
This experience report describes the design and implementation of a peer-led computer science TA training program that began in 2022 and has since undergone three iterations, inspired by the approach of agile software development. The current program consists of 10 sessions, organized to serve TAs in three respective stages of professional development. The iterations involved updating and enrichment of the syllabus, transitioning from lecture-centered to discussion-centered training, and discussions of emerging topics in computing education such as the use of large language models (LLMs). Participant feedback showed that TAs approved the iterative design of the training, while identifying areas for further improvement. We summarize lessons learned from the iterative process, reflect on the role of peer TA trainers, and discuss plans for future iterations.