Build literature surveillance agent using AWS Strands Agents SDK

Blog
August 22, 2025

Jayant Raj

Director, Cloud & Digital, AWS Ambassador, PwC US

AWS recently announced the release of Strands Agents SDK, an open-source SDK that simplifies the process of building and running AI agents using a model-driven approach. Strands Agents SDK relies on the advanced capabilities of state-of-the-art models to plan, chain thoughts, call tools and reflect, compared to other frameworks that need complex orchestration logic to build agentic workflows.

In this blog post, we will explore Strands Agents SDK by building and deploying an AI agent that monitors newly published scientific literature for mentions of specific drug and flagging potential safety signals or adverse events. Monitoring literature for adverse events is currently a very manual process. AI agents powered by reasoning of latest LLM models and tool-calling capabilities have great potential to automate the process.