Smart Feedback Loop
An AI-powered learning platform that delivers real-time, rubric-based critique to design students — closing the 72-hour gap between submission and meaningful feedback.
Title
Industry
Date
Challenge
Students in design courses submit work and wait 48–72 hours for feedback that arrives too late to influence their next iteration. When it does arrive, it's often generic, brief, and disconnected from specific design principles. Meanwhile, educators spend 6–8 hours per cohort writing near-identical comments — time that could be spent coaching rather than copy-pasting.
Approach
VisionVoice uses AI to generate rubric-aligned critique the moment a student submits — specific, principle-linked, and adjustable in tone. Educators shift from feedback producers to feedback curators, reviewing AI-generated notes and adding context where it matters. The system builds a visible record of student growth over time.
A 12-week end-to-end UX project — from user research through to a fully interactive hi-fi prototype and lightweight design system. Research spanned 8 interviews and a 42-person survey across design students and studio educators in Melbourne. Two rounds of usability testing, 28 Figma components, and one core design principle throughout: feedback should feel like a conversation, not a verdict.
Projects
Explore more like this one
Selected projects that reflect my approach to design, development, and execution.