Build Observable AI Agents with Strands, AgentCore, and Datadog

Learn to build production-ready AI agents with comprehensive operational visibility in this AWS re:Invent 2025 conference session. Walk through the complete process of constructing a Strands-based agent, deploying it on Amazon Bedrock AgentCore, and implementing telemetry processing with Datadog LLM Observability. Discover how to transform opaque large language model behavior into measurable and improvable workflows through end-to-end traces that span prompts, tool calls, and multi-agent interactions. Master techniques for improving response quality, tuning performance and cost optimization, and applying essential security and privacy guardrails. Gain practical best practices and access to a runnable reference implementation designed for production environments, presented by AWS Partner Datadog.