Feedbacker is an AI-assisted feedback tool designed to help educators provide high-quality, actionable feedback efficiently, without replacing human judgement. It supports formative assessment in higher education, helping staff manage workload while keeping feedback personalised, pedagogically grounded, and meaningful for students.
Origins
Feedbacker began as a personal solution to a common challenge: providing detailed, constructive feedback is time-intensive, and repetitive marking work can distract from the parts of teaching that really matter — pedagogy, conversations, and human connection with students. Dr Steve Huckle, as Module Convenor for Programming for 3D at the University of Sussex, developed an early version to maintain high-quality feedback at scale. The tool worked, and it genuinely helped.
Why Feedbacker Exists
As class sizes grow and assessment workloads increase, maintaining consistent, actionable feedback becomes challenging. Feedbacker helps educators scale thoughtful feedback while keeping academic expertise central. The platform combines human judgement with automation to streamline the repetitive aspects of marking, enabling personalised, actionable feedback more efficiently without compromising quality. Its aim is to free up time for teaching, discussion, and student engagement, while giving students a clearer understanding of their progress and next steps.
Responsible AI
Feedbacker generates draft feedback only. Educators remain fully in control, reviewing and editing all outputs. The system does not automate grading decisions, and all feedback is clearly attributable to the educator. This approach aligns with sector guidance in UK higher education, ensuring AI supports — rather than replaces — academic expertise.
Key Capabilities
- Rubric-driven assessment: Apply a single marking rubric consistently across individual or multiple student submissions.
- Batch feedback generation: Generate structured, criterion-referenced feedback for multiple submissions in one workflow.
- Human-in-the-loop design: Educators retain full control over rubric design, assessment criteria, and final judgements.
- Scalable feedback: Supports larger cohorts while maintaining clarity, pedagogical depth, and consistency.
- Transparency and consistency: Makes assessment criteria explicit and consistently applied.
- Flexible deployment: Open-licensed source code allows institutional hosting, adaptation, and integration.
Feedback Smarter, Faster, Better
Feedbacker helps educators provide high-quality feedback efficiently, keeping human judgement central.
Smarter
- Combines academic expertise with AI-assisted drafting
- Ensures feedback is pedagogically meaningful and actionable
- Maintains consistency across cohorts
Faster
- Automates repetitive aspects of marking
- Reduces turnaround time for student feedback
- Frees up time for teaching and discussion
Better
- Highlights student strengths and areas for improvement
- Points students toward clear next steps
- Supports formative, rubric-driven feedback workflows
Collaboration and Educational Context
Feedbacker is developed as part of ongoing exploration into responsible uses of AI in higher education assessment and feedback practices. The project welcomes discussion, experimentation, and collaboration with educators, learning technologists, and institutions interested in piloting or studying AI-assisted feedback workflows.
Getting Started in Your Context
Feedbacker can be used directly via the public deployment, or self-hosted by institutions. Self-hosting allows full control over data and AI configuration, making it suitable for pilot projects, module-specific trials, or integration with local assessment workflows. Educators remain fully responsible for reviewing and approving all outputs, ensuring alignment with institutional policies and pedagogical objectives.