Elizabeth Fuentes

Build, deploy, and scale agents with re-usable multi-agent patterns in Strands Agents. You will create an agentic system that is capable of decomposing complex tasks and delegating sub-tasks to specialized worker agents.

Voice agents powered by LLMs look amazing in demos… until you try to build one yourself and the problems start: slow responses, overlapping audio, and a bot that either hallucinates non-stop or falls into eternal silence.

The open-source Strands Agents SDK enables building production-ready, multi-agent AI systems in a few lines of code. The session demonstrates creating custom tools for video content analysis and converting them into Model Context Protocol (MCP) servers.

Modern observability extends beyond logs and metrics to deliver understanding, context, and prediction. This session demonstrates how Generative AI transforms DevOps and SRE practices in CI/CD environments.

In this session, see how you can build a WhatsApp assistant that uses Amazon Bedrock Agents, Retrieval Augmented Generation (RAG) using Knowledge Base with Amazon Aurora Serverless PostgreSQL-Compatible edition, and Amazon DynamoDB.

Discover how to build a real-time data visualization pipeline using Amazon OpenSearch Service, Amazon Redshift, and Amazon DynamoDB, without the complexity of traditional ETL processes.

What happens in Vegas should actually happen in the right Vegas! Dive into this chalk talk that unveils the magic of LangChain and RAG in transforming customer support.