Get Started Docs Learn Cloud Blog Use Cases Temporal Github Repository Sign Up for Cloud

Webinar

Building a Deep Research Agent on Amazon Bedrock Agentcore & Temporal

November 3, 2026 at 9am PT/12pm ET

In this workshop we will be Building a deep research agent that survives crashes. Learners will understand how Amazon Bedrock AgentCore supplies secure runtime, memory, browser and observability, while Temporal's durable execution handles retries, long waits and human-in-the-loop, so research resumes exactly where it left off.

We'll cover how a Temporal Workflow acts as the durable agent loop while an AgentCore Runtime session hosts the Temporal Worker. We'll also break down the architecture of a multi-agent research setup with isolated browser sessions, memory and observability, and show how it recovers from failures and resumes where it left off.

Key takeaways


  • Why durability matters for research agents: how long-running, multi-step agents recover from worker crashes without restarting from scratch
  • Division of responsibilities: AgentCore supplies runtime, memory, browser and code-interpreter sandboxes, and observability, while Temporal owns execution state, retries, timeouts and long waits
  • Anatomy of the architecture: Temporal Workflow as the agent loop, Serverless Worker on AgentCore Runtime, and isolated subagents researching in parallel
  • Human-in-the-loop and visibility: pausing for approvals or clarifying questions, and tracing execution with Temporal Event History alongside AgentCore Observability


Can't make the live session? Register anyway, and we'll send the recording to your inbox.

Presenter

Shubham Londhe

Senior Developer Advocate
Temporal Technologies

About the presenter

Shubham Londhe, is a Senior Developer Advocate at Temporal and a Google Developer Expert, passionate about developing and deploying production-ready applications with over 9 Years of Experience, with organizations like AWS, Liven, Vamstar UK.

He believes in continuous & Happy Learning to keep himself updated, as well as continues to share these learnings to over 350K subscribers across social media platforms with his initiative "TrainWithShubham".

YouTube Twitter LinkedIn Slack