Cloud Managed Agents (CMA) provide a scalable, infrastructure-managed framework for building, deploying, and orchestrating AI agents. The platform abstracts complex backend requirements like sandbox management, event streaming, and scaling, allowing developers to focus on product logic. Key capabilities include multi-agent delegation, where a primary agent coordinates sub-agents for specialized tasks, and persistent memory stores that enable collective learning across sessions. Security is managed through Vaults, which securely inject credentials into specific host domains, while "Dreaming" functionality allows agents to analyze past transcripts to refine future performance. By utilizing custom harnesses and asynchronous event-driven architectures, developers can build complex, long-running workflows—such as automated financial research—that operate autonomously, evaluate their own outcomes against defined rubrics, and continuously improve through shared knowledge and structured data extraction.
Sign in to continue reading, translating and more.
Open full episode in Podwise