AI-assisted formative feedback on image and video-based e-portfolio submissions

Apprenticeship completion rates stand at just above 50%, with higher dropouts at Level 3 and below. Of those who leave, 70% cite quality of training and lack of support as key reasons. Apprentices must submit portfolios for end-point assessments (EPAs), but feedback is often delayed, typically by 4–6 weeks. This delay makes it difficult for apprentices to track which knowledge, skills, and behaviours they’ve met and where gaps remain. The lack of timely feedback also results in missed opportunities for improvement, demotivation, reduced confidence, and strained relationships with apprentices, tutors and employers, ultimately leading to dropout.

This project will develop an AI-powered web app that delivers instant, personalised, examiner-level feedback on image and video portfolio submissions, which account for 30–40% of EPA evidence. By automatically mapping submissions to required competencies, the tool identifies demonstrated knowledge, skills, and behaviours and highlights areas for improvement. Immediate feedback will help apprentices focus their learning and enhance portfolio quality, while enabling tutors to offer targeted support.

By providing immediate feedback on multimedia evidence, this solution offers a new way to support skill development and behavioural demonstration, elements essential to effective vocational training.  This solution has the potential to improve completion rates among learners, creating broader impact across the apprenticeship sector.