Beyond Chatbots: How Amazon Bedrock Agents Drive Enterprise AI
summary
Claim: Amazon Bedrock Agents run managed, production-grade multi-agent workflows that orchestrate tools, knowledge bases and models with built-in observability. Demo builds an investment research assistant composed of three collaborators that query a Bedrock knowledge base of Amazon 10‑Ks, call web search and Yahoo Finance APIs, run analyses in a code sandbox and return analyst price ranges. The system executes a 43-step trace in ~3 minutes, supports memory, action groups and MCP connections, and produces concrete outputs like competitor lists and a $200–$261 analyst range.
AI-written summary · about
chapters (10)
- 0:00Introduction and Overview
- 0:22Understanding AI Agents
- 2:24The Rise of AI Agents
- 6:06AWS Generative AI Stack
- 8:15Amazon Bedrock Agents
- 15:02Challenges in Building AI Agents
- 17:03Amazon Bedrock Agents Capabilities
- 22:19Multi-Agent Collaboration
- 27:26Demo: Investment Research Assistant
- 42:17Conclusion and Contact Information
related talks
29mBuild and scale reliable multi-agent systems on AWS
38mCraft your own Generative AI Chatbot with Amazon Bedrock
30mAccelerate AI to prod
40mBuilding an AWS solutions architect agent with Amazon Bedrock
28mBuilding Agentic AI applications on AWS
42mExplore text-generation FMs for top use cases with Amazon Bedrock
24mBuilding Your Intelligent Data Agent
14mSkip the Middleman: Direct, Secure UI-to-Agent Communication with Amazon Bedrock AgentCore