Review the re:Inforce 2025 event catalog

This workshop takes a hands-on approach to generative AI security, focusing on Amazon Bedrock, Amazon SageMaker, and related services. We'll begin by examining Bedrock's core security principles, including data protection during inference and in features like agents, guardrails, and knowledge bases. Participants will gain insights into the internal architectures and security implications of context windows, system prompts, agent orchestration, and more. The session then transitions into hands-on red teaming exercises using SageMaker. We'll subsequently explore defensive strategies against these threat vectors and discuss methods for integrating these practices into development workflows. Participants will leave equipped with a holistic understanding of generative AI security, from individual model protection to safeguarding complex, multi-component systems. You must bring your laptop to participate.