Building secure AI agents at scale: Introducing Loom for AWS (aws.amazon.com)

Amazon Web Services (AWS) has introduced Loom for AWS, a robust platform designed to build and deploy secure AI agents at scale seamlessly. As organizations increasingly seek to leverage AI capabilities, Loom addresses the vital need for security and governance in building AI agents. It utilizes AWS Strands Agents and the Amazon Bedrock AgentCore Runtime to provide a low-code and no-code solution that simplifies agent deployment while enforcing enterprise-level security through features like role-based access control (RBAC) and comprehensive tagging for cost attribution. Loom stands out for its integration of enterprise best practices within its architecture, allowing platform engineering teams to create and manage AI agent applications efficiently. It ensures that agents operate with least privilege access and rigorous validation by utilizing a paved path blueprint approach for deployments. Furthermore, Loom’s use of a control plane API over AWS's existing SDK provides an opinionated layer that abstracts complexities, streamlining the deployment process. This new platform not only simplifies the creation and management of AI agents but also strengthens adherence to security protocols and governance, positioning it as a significant advancement for the AI/ML community focused on secure, scalable deployments.