Primary supervisor

Charith Jayasekara

Providing timely, individualised feedback is a persistent challenge in large-scale computing units.
This project investigates how Generative AI models can automatically produce pedagogically aligned, rubric-based feedback on student submissions.
A prototype system will interface with an LLM API (e.g., OpenAI GPT) and generate structured feedback, which will be evaluated for accuracy, usefulness, and tone against educator benchmarks.

Aim/outline

  • Design a feedback-generation pipeline linking uploaded student work to AI output.
  • Engineer prompts that ensure feedback relevance and academic integrity.
  • Implement a lightweight web interface for students and educators.
  • Evaluate generated feedback via qualitative and quantitative comparison.
  • Provide design recommendations for ethical AI feedback systems in higher education.

Required knowledge

  • Python (Flask or Streamlit preferred).
  • Basic understanding of LLMs and prompt engineering.
  • Familiarity with assessment rubrics and educational feedback principles.
  • Interest in applied AI and human-centred evaluation.