<![CDATA[Abstract This study examines automotive vocational teachers’ perceptions of the implementation of the deep learning approach at SMK Nasional Berbah and identifies internal and external factors associated with those perceptions. The study employed a descriptive quantitative design and was conducted from September to November 2025. Twelve productive automotive teachers in the Light Vehicle Engineering and Motorcycle Engineering programs were selected purposively. Data were collected through a 21-item questionnaire using a four-point Likert scale. The instrument covered internal factors, including attention, interest, and experience, and external factors, including knowledge, learning models, facilities and infrastructure, and school environmental support. The instrument had undergone construct validation through expert judgment and empirical testing, and reliability analysis produced a Cronbach’s alpha coefficient of 0.946. Data were analyzed using descriptive statistics, including percentages, means, and standard deviations, and were interpreted using predetermined categories. The results indicate that teachers generally had a positive perception of deep learning implementation, with 83% of respondents falling into the good category. Teachers considered the approach relevant for strengthening students’ understanding, learning engagement, and critical thinking. However, implementation was constrained by limited instructional time, administrative workload, facilities, and the need for more specific professional development. The findings suggest that successful implementation in vocational workshops requires not only teacher acceptance but also practical training, adequate facilities, and institutional support adapted to the characteristics of hands-on automotive learning. Abstrak Penelitian ini bertujuan mengetahui persepsi guru produktif otomotif terhadap implementasi pendekatan deep learning di SMK Nasional Berbah serta mengidentifikasi faktor internal dan eksternal yang berkaitan dengan persepsi tersebut. Penelitian menggunakan desain kuantitatif deskriptif dan dilaksanakan pada September–November 2025. Sebanyak 12 guru produktif pada program keahlian Teknik Kendaraan Ringan dan Teknik Bisnis Sepeda Motor dipilih secara purposive. Data dikumpulkan menggunakan kuesioner 21 butir dengan skala Likert empat tingkat. Instrumen mencakup faktor internal berupa perhatian, minat, dan pengalaman serta faktor eksternal berupa pengetahuan, model pembelajaran, sarana dan prasarana, dan dukungan lingkungan sekolah. Instrumen telah melalui validasi konstruk melalui expert judgment dan uji empiris, sedangkan pengujian reliabilitas menghasilkan koefisien Cronbach’s Alpha sebesar 0,946. Data dianalisis menggunakan statistik deskriptif berupa persentase, rerata, dan simpangan baku, kemudian diinterpretasikan berdasarkan kategori yang telah ditetapkan. Hasil penelitian menunjukkan bahwa persepsi guru terhadap implementasi deep learning cenderung positif, dengan 83% responden berada pada kategori baik. Guru menilai pendekatan ini relevan untuk memperkuat pemahaman, keterlibatan belajar, dan kemampuan berpikir kritis peserta didik. Meskipun demikian, implementasi masih menghadapi keterbatasan waktu pembelajaran, beban administrasi, kesiapan sarana-prasarana, dan kebutuhan peningkatan kompetensi guru. Temuan menunjukkan bahwa penerapan deep learning pada pembelajaran otomotif memerlukan dukungan pelatihan yang kontekstual, fasilitas praktik yang memadai, dan dukungan kelembagaan yang sesuai dengan karakter pembelajaran bengkel.]]>
