Agentic AI is transforming the way we build intelligent applications, but deploying AI agents securely and efficiently can be challenging. This tutorial demonstrates how to deploy an AI agent built with Google's Agent Development Kit (ADK) to Cloud Run using GitLab's native integrations and CI/CD components. AI agents leverage advanced language models and natural language processing to take independent action, making decisions, and executing multistep plans to achieve goals autonomously. The tutorial uses a practical example, the Canada City Advisor, which helps users find their ideal Canadian city based on their preferences and constraints. To deploy the AI agent, you need a Google Cloud project with the Cloud Run API, Artifact Registry API, and Vertex AI API enabled, as well as a GitLab project for your source code. The first step is to set up IAM integration with Workload Identity Federation, which eliminates the need for service account keys and improves security. Next, you need to configure Google Artifact Registry integration and add necessary roles to your service principal for Cloud Run deployment. The CI/CD pipeline is then created using a .gitlab-ci.yml file, which consists of four stages: build, test, upload, and deploy. The pipeline provides several security advantages, including no long-lived credentials, automated security scanning, an audit trail, and the principle of least privilege. By combining GitLab's security features with Google Cloud's powerful AI and serverless platforms, you can deploy AI agents that are both secure and scalable.