Building AI Agents with Strands Agents and Amazon Bedrock

The Strands Agents SDK is an open-source framework for building AI agents in just a few lines of code. Developers define a set of tools and a system prompt, and the underlying model handles reasoning, planning, and execution. Custom tools are registered with a single decorator, and the SDK manages the full agentic loop from invocation to response.

Amazon Bedrock is a fully managed service that provides access to high-performance foundation models from Amazon and leading AI providers. The Amazon Nova Lite foundation model is a fast, cost-efficient model from Amazon's Nova family that supports tool use and multi-step reasoning out of the box.

In this lab, you will build and deploy an agent powered by the Strands Agents SDK to AWS Lambda, implement custom tools that the agent can reason over and chain together, and observe the full agentic loop.

Learning objectives

Upon completion of this beginner-level lab, you will be able to:

• Deploy an AI agent using the Strands Agents SDK and Amazon Bedrock
• Integrate the AI agent with various tools and APIs to perform complex tasks
• Observe and understand the agentic loop in action, including reasoning, planning, and execution

Intended audience

• Candidates for the AWS Certified Generative AI Developer – Professional (AIP-C01) exam
• Cloud Architects
• AI Developers

Prerequisites

Familiarity with the following will be beneficial but is not required:

• Amazon Bedrock
• AWS Lambda

The following content can be used to fulfill the prerequisites:

• Streamlining Workflows and Tasks with Amazon Bedrock Agents
• Working With Conversational Memory and the Amazon Bedrock Converse API