Software

Amazon Introduces Amazon Bedrock AgentCore for AI-Driven Development

Amazon Bedrock AgentCore enables AI to assist in software development workflows, with reference implementations showing how it automates ER diagram generation and code security analysis.

"Digitalization for decommissioning projects (011111693)" by IAEA Imagebank is licensed under CC BY 2.0. To view a copy of this license, visit https://creativecommons.org/licenses/by/2.0/.

Amazon announced Amazon Bedrock AgentCore, a service for building and optimizing agents at scale, on its blog. It is the company's latest software update in the machine learning space, following its previous release of Amazon Bedrock in 2023.

Amazon Bedrock AgentCore allows developers to create agents that can assist with tasks like generating Mermaid entity relationship diagrams from SQL schemas and performing automated code security analysis. The service supports multiple frameworks and models, including Kiro and Claude Sonnet 4 through Amazon Bedrock.

"AI-DLC positions AI as a central collaborator across the software development lifecycle, handling routine execution while humans retain oversight of critical decisions," said the blog post. This approach aims to maintain human-in-the-loop governance while accelerating delivery.

The announcement follows Amazon's broader push to integrate AI into software development processes, with the company emphasizing the importance of structured workflows and modular components. Amazon Bedrock AgentCore is designed to help teams compose AI-driven workflows that maintain governance while improving efficiency.

Amazon did not specify the exact number of models supported by AgentCore, and the service's full capabilities are still being explored. The company plans to release more reference implementations in the coming months, with details available in their GitHub repositories.

Source: awsml

Key points

Amazon Bedrock AgentCore is a service for building, connecting, and optimizing agents at scale with any framework or model.
AI-DLC positions AI as a central collaborator across the software development lifecycle, handling routine execution while humans retain oversight of critical decisions.
The first reference implementation generates Mermaid entity relationship diagrams from SQL schemas using Amazon Bedrock AgentCore.
The second reference implementation provides automated code security analysis through a multi-agent architecture using Amazon Bedrock AgentCore.
Both implementations link to complete deployment instructions in their respective GitHub repositories.
The system uses a serverless, event-driven architecture with components including Amazon S3, AWS Lambda, and Amazon Cognito for authentication.
The agent implementation uses the BedrockAgentCoreApp runtime wrapper with the .entrypoint decorator to register the handler.

Source: AWS Machine Learning Read the original →

WRITTEN BY

Theo Almeida

AI Software & Developer Tools

Theo covers AI software, developer tools, frameworks, and the platforms builders use every day.