The AgentVerse - Build fast, Scale smarter
summary
Demonstrates building Amazon Bedrock agents that use Claude 3.5 Haiku or Amazon Nova, embeddings stored in vector stores (Titan or pgvector) and Lambda-backed API actions to create a shopping assistant that places an order end-to-end in 24 seconds. Shows multi-agent collaboration with a supervisor routing to product discovery and order-management agents, defaulting to five retrieved docs, and traceable chain-of-thought execution. Cites Gartner (by 2028 one‑third of apps agentic, 15% decisions autonomous) and claims up to 5% sales lift and models up to 75% lower cost.
AI-written summary · about
chapters (13)
- 0:00Introduction
- 1:28AI agents and business impact
- 4:21Production challenges
- 6:07Amazon Bedrock simplifies
- 7:56Building with Bedrock Agents
- 10:44AI agents use cases
- 12:50Demo shopping assistant
- 21:19Multi-agent collaboration
- 27:59Session state for context
- 31:07Operational efficiency
- 32:55HR support demo
- 38:17Scaling securely
- 40:42Key takeaways
related talks
19mBuilding Agents at Cloud Scale
28mBuilding Agentic AI applications on AWS
21mHarnessing Technological Innovation & Accelerating Uptake
49mBuilding AI Agents with Spring & MCP
40mAI Agents in Practice
14mThe New Web Is Agentic. Is Your Infrastructure Ready?
43mArchitecting the Future: Modern Solution Design with Agentic AI
49mAI Agents: Collaboration in Real-Life Applications