GenAI-Driven Knowledge Bases: Optimizing AWS Support and Cloud Ops
summary
A working GenAI pipeline consolidates AWS Support cases and AWS Health notifications into Amazon Bedrock via JSON stored on S3 ingestion and nightly EventBridge sync, uses Lambda (assume-role across accounts deployed by CloudFormation StackSet) to call describe-events/describe-cases, and exposes a Bedrock knowledge base with RAG retrieval and prompt templates. Demo returns summaries, support recommendations, ARNs, step-by-step remediation (example: Python 3.8 Lambda EOL), and notes a document-return limit of up to 100 and recommends Athena for large-scale analytics and metadata filters.
AI-written summary · about
chapters (8)
- 0:00Introduction to GenAI knowledge bases
- 2:05Virtual assistant for AWS support
- 3:28GenAI stack at AWS
- 5:15Solution architecture overview
- 6:39Demo of knowledge base queries
- 13:44Enhancing the solution
- 20:19Agentic use cases and future steps
- 22:55Q&A and closing remarks
related talks
16mEnterprise AI Innovation: GenAI Experimentation
29mCustomizing genAI applications for your business
30mStreamlining data governance
16mEnterprise AI Innovation: Gen AI Experimentation
01h 00mSecurely accelerating generative AI innovation
43mBeyond Chatbots: How Amazon Bedrock Agents Drive Enterprise AI
20mBuilding a secure end-to-end generative AI application in the cloud
40mBuilding an AWS solutions architect agent with Amazon Bedrock