Fei Yuan

AI agents act on behalf of users and systems, and delegating the right access without over-permissioning is hard. In this chalk talk, we use code examples to work through authorization delegation with Amazon Bedrock AgentCore Identity.

AI agents call APIs, access databases, and act on behalf of users, making security non-negotiable before you go to production. But they introduce new attack surfaces: persistent sessions, autonomous tool execution, and permissions that compound across steps.

As generative AI agents take on more complex tasks, the need for secure, contextual identity becomes critical. In this chalk talk, dive into Amazon Bedrock AgentCore Identity, the foundational building block that enables secure, auditable, and role-aware execution for AI agents.

As enterprises rapidly adopt AI agents, managing agent identities across environments becomes a critical challenge. SailPoint, a leading identity security ISV serving 85% of Fortune 500 companies, is leveraging Amazon Bedrock AgentCore to transform their product into a comprehenhensive AI agent governance solution.

Machine learning involves more than just training models; you need to source and prepare data, engineer features, select algorithms, train and tune models, and then deploy those models and monitor their performance in production. Learn how Amazon Consumer Payments uses Amazon SageMaker, AWS Glue, AWS Step Functions, Amazon API Gateway, AWS Lambda, AWS Batch, Amazon Elastic Container Registry...

Machine learning involves more than just training models; you need to source and prepare data, engineer features, select algorithms, train and tune models, and then deploy those models and monitor their performance in production. Learn how Amazon Consumer Payments uses Amazon SageMaker, AWS Glue, AWS Step Functions, Amazon API Gateway, AWS Lambda, AWS Batch, Amazon Elastic Container Registry...