Dive into AI security hands-on! Learn how to protect prompts with Model Armor and automate container scans using GeminiCLI + Trivy. From VertexAI Workbench to CI/CD pipelines — discover how to keep your AI projects safe, smart, and deployment-ready
0 RSVP'd
Get tickets
About the Session:
AI is powerful — but with great power comes great responsibility. Join us for this hands-on pre-DevFest workshop where we dive deep into securing AI workflows using GeminiCLI, Model Armor, and Trivy on Google Cloud.
In this session, you’ll explore how to protect prompts, automate container image scanning, and integrate security directly into your CI/CD pipelines — all while gaining real-world experience with the latest AI safety tools.
What You’ll Learn:
Prompt Sanitisation: Discover how Model Armor on VertexAI Workbench helps secure input and output prompts against injection and data exposure risks.
Secure Deployment Pipelines: Learn to use GeminiCLI and Trivy to automate container image scanning on a GCE VM, ensuring safer, cleaner deployments.
Why Attend:
Whether you’re an AI developer, a cloud enthusiast, or a security-minded builder, this session will help you integrate best practices for AI model protection, deployment hygiene, and trustworthy automation.
Get ready to roll up your sleeves, experiment with live demos, and walk away equipped to make your AI projects safe, scalable, and deployment-ready.
Saturday, December 20, 2025
4:00 AM – 11:30 AM (UTC)
| Registration |
| Openning Keynote |
| Input and Output prompt sanitisation via Model Armor demonstrated on VertexAI Workbench |
| Networking and Lunch |
| Using GeminiCLI and Trivy on GCE VM to automate container image scan before deployment as part of a CICD Pipeline example |
| Swag Distribution and End Note |
