Optimalisasi Kompetensi Guru melalui Workshop Deep Learning di Sekolah Dasar

Abstract

The implementation of deep learning as a teaching approach in the Merdeka Curriculum requires elementary school teachers to possess strong pedagogical competencies and digital literacy. However, most teachers in the Bima Cluster, Batealit Subdistrict, Jepara Regency have not yet received training on deep learning. This initiative aims to enhance teachers’ competencies through workshops and guidance on implementing deep learning. The program was attended by 45 teachers from six elementary schools, with activities focused on reinforcing deep learning concepts and integrating AI into instruction. The methods used included interactive lectures, discussions, hands-on development of teaching materials, learning simulations, and mentoring. Evaluation was conducted using pretests, posttests, observations, and participant feedback surveys. The results showed that the average teacher comprehension score increased, with an N-Gain value of 0.60 (moderate category), and the average participant response was 4.84 (very good category). These improvements indicate that teachers not only gained conceptual understanding but were also able to design more contextual, learner-centered instruction and utilize AI to support the development of instructional materials and authentic assessments. These findings indicate that practice-based training models are effective in enhancing teachers’ readiness to implement deep learning in the classroom. The sustainability of the program through teacher learning communities, mentoring, and ongoing training is necessary so that changes in competencies can be internalized over the long term and lead to improvements in the quality of learning in elementary schools.