“Introduction to Statistical Methods” and “From Molecules to Ecosystems” are first-year units in the School of Life and Environmental Sciences with large enrolments up to 1400 students. Large classes pose challenges in providing personalised, positive, and timely assessment feedback due to heavy workloads, the need for consistency, and budget constraints when delegating marking to casual staff. To address this, we built an AI assistant using Cogniti, a generative AI tool created by educators at the University of Sydney. Cogniti allows educators to steer large language models (LLMs) like GPT-4 with instructions and resources, producing tailored outputs that support student learning. The Cogniti agents, designed as either a biology expert or data scientist, aid in delivering constructive feedback on written assessments while maintaining marker control over the grading process. We describe the impact of using Cogniti as a feedback expander, demonstrate the types of feedback it provides, and outline its adaptability for new units of study.
Teaching@Sydney