Building GenAI applications with Amazon Bedrock
summary
Overview of building production GenAI on AWS with concrete details: uses Amazon Bedrock for LLM inference (Nova family, Anthropic Claude, Titan, newly added DeepSeek R1), configures Amazon Bedrock agents with Lambda tools, hierarchical multi‑agent orchestration and visible Chain of Thought trace, and connects private or public datasets via knowledge bases backed by OpenSearch Serverless, Aurora PGVector, Pinecone or S3 ingestion. Demoed Eliza OS integration, a text‑to‑SQL pipeline over a seven‑chain public blockchain dataset, and console‑first agent setup that runs without custom infra.
AI-written summary · about
chapters (11)
- 0:00Introduction & Background
- 1:18Amazon Bedrock Overview
- 2:29Anthropic Claude & Amazon Titan
- 3:48Amazon Nova Foundation Models
- 5:06Multi-Agent Collaboration
- 6:20Setting Up Bedrock Agents
- 7:33Knowledge Bases & Rag
- 9:26Data Ingestion & Vector Stores
- 10:35Public Blockchain Dataset on AWS
- 11:52Integrating Bedrock with Eliza OS
- 13:04Getting Started & Further Resources
related talks
16mEnterprise AI Innovation: Gen AI Experimentation
29mCustomizing genAI applications for your business
16mEnterprise AI Innovation: GenAI Experimentation
38mGenerative AI - Architectures and applications in depth
55mSecurely build generative AI apps & control data with Amazon Bedrock
43mBuilding an AI-Ready Data Foundation
01h 30mUnleashing the Power of AWS: A Workshop on Bedrock Assistant
48mHow to Make Smart Architecture Decisions when Building Gen AI Apps