AI grading and feedback, insights from a programming course - Luke Korthals (UvA) скачать в хорошем качестве

AI grading and feedback, insights from a programming course - Luke Korthals (UvA)

In this presentation, Luke Korthals (UvA) shares insights from a real-world experiment in a programming course, where GPT-4o was used to grade open-ended code submissions and generate personalised feedback. Using a setup that combined Canvas and a Shiny app, the approach reduced grading time and costs while improving feedback quality. Korthals also discusses key challenges, including rubric design, alignment, and the difference between subtractive and positive scoring methods. This video was recorded during the event Grading exams with generative AI (VU CTL, 2025). At this event, experts shared insights, experiences, and critical reflections on the use of generative AI in exam grading. Read more about it here (Dutch): https://community-toetsen.npuls.nl/gr... To see more presentations from this event, explore the full playlist. To stay informed about how AI is shaping education, subscribe to this channel.