Claim Missing Document
Check
Articles

Found 13 Documents
Search

IMPLEMENTATION OF VIBE CODING IN THE DEVELOPMENT OF AN INTERNAL QUALITY ASSURANCE MANAGEMENT SYSTEM FOR INTERNAL QUALITY AUDIT (AMI) AND PPEPP BASED ON LAM INFOKOM STANDARDS Ahmad Syaifuddin; Ronny Makhfuddin Akbar; Hafizh Fianto Putra; Yanuarini Nur Sukmaningtiyas; Anita
IJCDE (Indonesian Journal of Community Diversity and Engagement) Vol. 7 No. 1 (2026): Vol. 7 No. 1 , 2026
Publisher : LEMBAGA PENELITIAN, PENGABDIAN PADA MASYARAKAT, PENINGKATAN AKTIVITAS INSTRUKSIONAL, PENINGKATAN DAN PENJAMINAN MUTU UNIVERSITAS ISLAM MAJAPAHIT

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

The Internal Quality Assurance System (IQAS), implemented through the Establishment, Implementation, Evaluation, Control, and Improvement (PPEPP) cycle, is a mandatory requirement for study programs seeking accreditation from the Independent Accreditation Agency for Informatics and Computer Science (LAM INFOKOM). Conventional management of Internal Quality Audit (AMI) documents and Self-Evaluation Reports (LED) often hinders accreditation preparation due to fragmented, inconsistently archived documentation. This community service activity assisted the Quality Assurance Unit at Universitas Islam Majapahit (UNIM) in developing an integrated quality management information system using Native PHP and Tailwind CSS through a vibe coding approach, in which developers with existing web development competence directed an AI programming assistant to generate source code and interface components. The system was completed as a functional prototype within one month, faster than typical conventional development. Following Focus Group Discussion sessions and hands-on mentoring, self-assessed readiness among five participating Quality Assurance Unit members improved across all evaluated aspects. These findings suggest that AI-assisted development, guided by developers with relevant technical competence, offers a practical and time-efficient pathway for institutions with limited programming resources to build internal quality assurance systems aligned with LAM INFOKOM standards.
Pendampingan Pengembangan Sistem Informasi AMI Berbasis PPEPP dengan Vibe Coding pada Program Studi Informatika UNIM Ahmad Syaifuddin; Ronny Makhfuddin Akbar; Hafizh Fianto Putra; Yanuarini Nur Sukmaningtiyas; Anita
Jurnal Bakti Dirgantara Vol. 3 No. 2 (2026): Jurnal Bakti Dirgantara (In-Press)
Publisher : Universitas Dirgantara Marsekal Suryadarma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35968/jdqrbx63

Abstract

Implementing the Internal Quality Assurance System (SPMI) via the PPEPP cycle is a mandatory requirement for study programs to align with LAM INFOKOM accreditation criteria. The assessment process is often hindered by the management of the Self-Evaluation Report (LED) and Internal Quality Audit (AMI) documentation, which still rely on manual, conventional filing methods. To address these issues, this community service initiative was organized to assist the quality assurance team in designing an integrated management information platform. The innovative approach adopted for this software development is "vibe coding" leveraging artificial intelligence as a virtual assistant to accelerate code syntax generation. The system's data architecture and logic were developed using a PHP framework, while the user interface design utilized the flexibility of Tailwind CSS. The project outcomes confirm that the application of vibe coding drastically reduced the time required to build the application. This system enables the quality assurance team to conduct simulated assessments using LAM INFOKOM instruments much more efficiently, a success reflected in a high user satisfaction index of 94.4%. Furthermore, the series of technical assistance sessions, combined with Focus Group Discussions (FGDs), proved effective
Implementasi dan Evaluasi Aplikasi Text-To-Image Berbasis Stable Diffusion Menggunakan Streamlit Dengan Optimasi Parameter Generasi Citra Ronny Makhfuddin Akbar; Yanuarini Nur Sukmaningtyas
SUBMIT: Jurnal Ilmiah Teknologi Infomasi dan Sains Vol. 6 No. 1 (2026): Juni 2026
Publisher : Program Studi Teknik Informatika, Universitas Islam Majapahit Mojokerto, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36815/submit.v6i1.4790

Abstract

Perkembangan Generative Artificial Intelligence (Generative AI) mendorong pemanfaatan teknologi text-to-image untuk menghasilkan citra digital berdasarkan deskripsi teks. Salah satu model yang banyak digunakan adalah Stable Diffusion karena mampu menghasilkan citra berkualitas tinggi dengan kebutuhan komputasi yang relatif efisien. Namun, kualitas citra sangat dipengaruhi oleh konfigurasi parameter inferensi sehingga diperlukan evaluasi untuk memperoleh konfigurasi yang optimal. Penelitian ini bertujuan mengimplementasikan dan mengevaluasi aplikasi Text-to-Image berbasis Stable Diffusion menggunakan Streamlit serta menganalisis pengaruh parameter CFG Scale, Inference Steps, dan Scheduler terhadap kualitas citra dan efisiensi inferensi. Penelitian menggunakan metode eksperimen dengan pendekatan One Factor at a Time (OFAT). Pengujian dilakukan menggunakan lima text prompt yang mewakili kategori Human Portrait, Animal, Nature, Architecture, dan Fantasy. Evaluasi dilakukan menggunakan metrik Inference Time, Image Entropy, Brightness, Contrast, dan Image Size. Hasil penelitian menunjukkan bahwa CFG Scale 7.5 menghasilkan keseimbangan terbaik antara kualitas visual dan naturalitas citra, Inference Steps 30 memberikan kualitas citra yang baik dengan waktu inferensi yang efisien, sedangkan Scheduler DPM++ menghasilkan performa terbaik berdasarkan kompleksitas informasi visual dan kontras citra. Kombinasi ketiga parameter tersebut direkomendasikan sebagai konfigurasi optimal untuk menghasilkan citra berkualitas dengan efisiensi komputasi yang baik pada aplikasi Text-to-Image berbasis Stable Diffusion.