AWS Launches Strands Labs for Agent-Based AI
AWS launched Strands Labs, a GitHub organization for experimental projects centered on agent-based AI, signaling a growing toolkit for automating parts of the ML pipeline.
Strands Labs includes projects like Robots, Robots Sim, and AI Functions, each exploring different facets of agent development, from robotics to code generation. The Robots project, for instance, focuses on connecting AI agents with physical hardware using a unified interface. AWS collaborated with NVIDIA, integrating the NVIDIA GR00T model into Strands Agents, showcasing sophisticated AI executing directly on embedded systems. They also integrated with Hugging Face's LeRobot, which simplifies interaction with robotics hardware and datasets. The Strands Agents SDK has gained traction, with over 14 million downloads since its open-source release in May 2025. This SDK, available in Python and TypeScript, allows developers to build AI agents using a model-driven approach. By separating experimental projects into Strands Labs, AWS allows teams across Amazon to contribute innovative open-source initiatives for community testing. This separation ensures that experimental features undergo thorough community testing before potential integration into the main SDK.
