AI article
Building Production-Ready Open-Weights AI Agents with Gemma, Google ADK, and Cloud Run GPUs
Community description: How I designed, connected, scaled, governed, evaluated, and QLoRA fine-tuned an enterprise SecOps & FinOps agent on Google Cloud.
Dev.to | Oct 16, 2026 | Robina Mirbahar
Automated excerpt
Figure 2: End-to-end request lifecycle for my CloudGuard Incident & FinOps Agent across Cloud Run, vLLM, Cloud Firestore, Model Armor guardrails, and Cloud Trace. A Google Cloud Project with billing enabled. Native Firestore & FinOps Tools: cloudguard_agent/tools. py Model Armor & RBAC Guardrails: cloudguard_agent/guardrails. py ADK Eval & OpenTelemetry Runner: scripts/run_eval. py Google Gemma Open Models Overview: ai. google. dev/gemma Google Agent Development Kit (ADK) Documentation: google. github. io/adk-docs Deploying Gemma on Vertex AI Model Garden: cloud. google. com/vertex-ai/generative-ai/docs/open-models/use-gemma Cloud Run Serverless GPUs Documentation: cloud. google. com/run/docs/configuring/services/gpu Cloud Storage FUSE Volume Mounts on Cloud Run: cloud. google.
Selected automatically from source text; not independently written or fact-checked. Read the original for full context.