TLDR: AWS has launched a suite of new agentic AI tools and services, including Amazon Bedrock AgentCore, Amazon S3 Vectors, and the Kiro agentic IDE, to enable the secure and scalable deployment of AI agents. This initiative is backed by a $100 million investment in the Generative AI Innovation Center and aims to transform software development and enterprise productivity.
Amazon Web Services (AWS) has made a significant stride in the artificial intelligence landscape with a comprehensive unveiling of new agentic AI capabilities and development tools at the AWS Summit New York 2025. These innovations are poised to redefine how AI agents are built, deployed, and managed, emphasizing security, reliability, and scalability for enterprise applications. The announcements, spearheaded by Swami Sivasubramanian, AWS VP for Agentic AI, underscore a ‘tectonic change’ in software development and interaction.
At the forefront of these new offerings is Amazon Bedrock AgentCore, a robust suite of seven core services designed to facilitate the deployment and operation of highly effective AI agents across any framework and model. This addresses the critical challenge organizations face in moving AI agent prototypes to production, ensuring they meet stringent enterprise-grade security and governance standards. Key components of AgentCore include:
AgentCore Runtime: Provides a secure and dynamic environment capable of handling interactive experiences with low latency and complex asynchronous workloads lasting up to eight hours, offering industry-leading session isolation.
AgentCore Memory: Enables developers to build context-aware agents with superior long-term and short-term memory accuracy.
AgentCore Identity: Ensures seamless and secure agent authentication through integration with established identity providers such as Amazon Cognito, Microsoft Entra ID, and Okta.
AgentCore Gateway: Offers a secure mechanism for agents to discover and utilize a wide array of tools, transforming existing APIs, Lambda functions, and services into agent-compatible resources.
AgentCore Code Interpreter: Allows AI agents to securely write and execute code within customizable sandbox environments for tasks like complex calculations, reasoning validation, data processing, and visualization.
AgentCore Browser Tool: A cloud-based browser that empowers AI agents to interact with websites at scale for tasks such as form completion and navigation.
AgentCore Observability: Leverages Amazon CloudWatch to provide real-time visibility into agent performance through built-in dashboards and telemetry, integrating with existing observability systems.
Further solidifying its commitment to agentic AI, AWS announced an additional $100 million investment in the AWS Generative AI Innovation Center. This funding aims to accelerate the development and deployment of autonomous AI systems, building upon two years of successful customer empowerment in boosting productivity and transforming experiences. Companies like Warner Bros. Discovery Sports Europe, BMW, Syngenta, and AstraZeneca are already leveraging agentic AI on AWS for transformative results.
To simplify enterprise AI agent adoption, AWS has introduced a new ‘AI Agents and Tools’ category in AWS Marketplace. This centralized hub allows customers to easily discover, purchase, deploy, and manage AI agents and tools from leading providers, streamlining AI initiatives and accelerating development with ready-to-integrate solutions.
Enhancements to Amazon Nova models on Amazon SageMaker AI now offer advanced customization capabilities, supported by SageMaker HyperPod. This enables customers to fine-tune Nova models for specific needs with greater accuracy and flexibility. The Nova Act SDK has also been updated with improved security and access controls, facilitating the transition of prototypes to production-ready agents capable of web-based actions.
A significant infrastructure innovation is the preview launch of Amazon S3 Vectors, the first cloud object storage with native vector support for AI workloads. This service promises to reduce the cost of storing and querying vector data by up to 90% compared to traditional methods, making it highly cost-effective for retaining large vector datasets. It integrates seamlessly with Amazon Bedrock Knowledge Bases and OpenSearch Service, optimizing Retrieval Augmented Generation (RAG) and vector search operations.
For developers, AWS introduced Kiro, a new agentic Integrated Development Environment (IDE). Kiro streamlines the development workflow for AI agents by combining natural language prompting with established software engineering practices. It facilitates the generation of scalable and maintainable code, transforming static documents into ‘living, self-documenting sources of truth,’ and incorporates ‘Agent Hooks’ for event-driven automation.
Amazon Q Developer is also evolving to become more agentic, offering expanded Model Context Protocol (MCP) support, real-time code suggestions across over 25 languages, automated multi-file features, and intelligent code reviews. The service has already demonstrated substantial growth, processing millions of user interactions and tool calls daily.
Additionally, AWS announced updates to the open-source SDK Strands Agents 1.0, which dramatically simplifies the creation of multi-agent AI systems. This new developer tool reduces the complex technical work previously required over months to just hours, enabling businesses to build coordinated teams of AI assistants for tasks like customer service and data analysis.
Also Read:
- SuperOps and AWS Introduce Agentic AI Marketplace to Revolutionize MSP Operations
- ShellKode and AWS Forge Strategic Alliance to Propel Generative AI Adoption Across Indian Enterprises
Swami Sivasubramanian emphasized the foundational role of vectors in AI, stating, ‘vectors are the language of AI, capturing context and relationships within data, storing memory, retrieving relevant information, and correlating insights across petabytes of data.’ These comprehensive announcements position AWS as a leader in the rapidly evolving field of agentic AI, providing a robust ecosystem for developers and enterprises to build dynamic and intelligent applications.
