End-to-end GCP IoT analytics platform ingesting 10GB/day from 50,000+ devices through Cloud IoT Core and Pub/Sub into Dataflow streaming pipelines. Real-time telemetry flows through validation, enrichment, and 5-minute windowing before landing in BigQuery, while Cloud Composer orchestrates hourly Spark batch jobs on Dataproc for feature engineering. Vertex AI handles the complete ML lifecycle from training on BigQuery features to serving predictions via auto-scaling endpoints with minimum 2 replicas. Fork this architecture on Diagrams.so to customize the streaming windows, adjust Dataproc cluster sizing, or swap in your own ML model deployment strategy.