Agent Platform and Generative AI Security
In this 2-day course we’re taking a look at how to adopt Google’s Agent Platform and generative AI safely, with security at the center.
The Agent Platform and Generative AI Security course gives AI practitioners and security engineers targeted knowledge and hands-on skills, through eight modules and ten labs, to build a security-conscious approach to integrating gen AI. You will start with the foundations of Agent Platform and its security challenges, then implement identity and access control to restrict access to Agent Platform resources, configure encryption and protect sensitive information, and secure endpoints and model deployment.
You will enable logging, monitoring and alerting for real-time oversight, identify and mitigate the security threats that are unique to generative AI, and apply testing techniques to validate and secure model responses. You will also learn best practices for securing data sources and responses in Retrieval-Augmented Generation (RAG) systems, and build foundational knowledge of AI Safety. Labs include service accounts and roles, Cloud KMS, creating a de-identified copy of data, and safeguarding with the Agent Platform Gemini API.
The course is aimed at AI practitioners, security professionals and cloud architects. You should have fundamental knowledge of machine learning, in particular generative AI, and a basic understanding of security on Google Cloud.