Building Your Intelligent Data Agent
summary
Demo shows an agent that translates natural language to SQL using Amazon Bedrock Structured Knowledge Bases then runs queries against Amazon Redshift and Data Lakes, returning the same results (example: 6,842 orders / 340 cancellations). The system enforces role-based PII gating, applies user credentials to SQL before retrieval, logs model invocations and audit trails, and claims SOC2/HIPAA-compliant controls. Architecture combines Bedrock agents for NL→SQL, curated schema/context to prevent syntax hallucinations, and rate limiting to control cost and latency.
AI-written summary · about
chapters (9)
- 0:00Introduction and welcome
- 0:36Business data challenges today
- 2:03Story of Lucy and Jake
- 4:51What is an intelligent data agent
- 5:56Demo of the agent in action
- 10:20Agent architecture explained
- 17:23How agents work with permissions
- 18:25Best practices for the system
- 22:45Happy ending and resources
related talks
47mAgentic AI with Commercial Databases
43mBeyond Chatbots: How Amazon Bedrock Agents Drive Enterprise AI
40mData Intelligence Platform on AWS
30mBuilding a Self-Service Platform for Stateful AI Agents
40mBuilding Intelligent Search Systems With Agentic RAG and OpenSearch
19mDatabricks
48mHow to Make Smart Architecture Decisions when Building Gen AI Apps
19mBuilding Agents at Cloud Scale