1. Mengapa Pelatihan Custom Extractor di Google Cloud Document AI Gagal Mengakses Bucket GCS?
Saat membangun pipeline ekstraksi dokumen otomatis sebagai GCP Professional Data Engineer dan AI Application Engineer menggunakan Google Cloud Document AI Workbench, proses impor dataset dan pelatihan model sebenarnya dieksekusi oleh Google-managed Service Agent internal (service-PROJECT_NUMBER@gcp-sa-prod-dai-core.iam.gserviceaccount.com), bukan oleh akun pengguna atau Service Account aplikasi Anda.
2. Konfigurasi IAM Least-Privilege pada Google Cloud Storage untuk Document AI
Untuk memperbaiki kegagalan akses bucket GCS saat membaca dokumen PDF berlabel dan menulis artefak evaluasi model, aktifkan tampilan Include Google-provided role grants di konsol IAM atau berikan peran roles/storage.objectAdmin melalui Terraform/gcloud CLI seperti pada panduan berikut:
{
"status": {
"code": 7,
"message": "Does not have permissions [storage.objects.get] in the bucket"
}
}
Ringkasan Arsitektur Produksi & Praktik Terbaik (Checklist Teknis)
Keandalan & Observabilitas Produksi: Setiap implementasi wajib dilengkapi metrik latensi P50/P95/P99, penanganan kegagalan idempoten, dan pengujian otomatis di CI/CD.
Keamanan & Isolasi Least-Privilege: Validasi seluruh input eksternal, gunakan enkripsi standar industri, dan terapkan prinsip hak akses minimum pada layanan cloud maupun penyimpanan lokal.
About the Author — Hire Aldian Fazrihady (Cloud Engineer, GCP Professional Cloud Architect, GCP Professional Data Engineer, Backend, AI & Full Stack Engineer)
Aldian Fazrihady is a Top-Rated Senior Cloud Engineer, GCP Professional Cloud Architect (Google Cloud Certified), GCP Professional Data Engineer (Google Cloud Certified), Backend Engineer (Python, Django, FastAPI, Go, PostgreSQL), AI Application Engineer (LLMs, RAG, MCP, Google ADK, LangGraph), and Full Stack Engineer (React, Next.js, TypeScript) with 15+ years of production software engineering experience.