📙 Incident Readiness You and Your Leaders Will Actually Trust 🔥
summary
A repeatable process called an environmental intelligence report queries existing logs (Athena, Splunk, S3), stores queries in notebooks, and uses LLM skills and sub-agents (Claude via Bedrock recommended) to produce templated, citation-backed findings with validation and hedging. Real results include dormant admins, MFA gaps, deprecated policies, service-account misuse, and cost leaks; six customers saved multi-million dollars in cloud spend. The pipeline scales (over 10+ petabytes/day processed at prototype scale) and is presented as a productized capability ready to run weekly for measurable remediation.
AI-written summary · about
chapters (7)
- 0:00Introduction and speakers
- 0:42Why visibility in production is hard
- 3:34Environmental intelligence reports
- 10:40Generating reports with LLMs
- 15:14Driving action from reports
- 19:08Key takeaways and recommendations
- 20:58Q&A session
related talks
15mIncident Ready: Chaos Engineer Your Incident Response
36mModernizing Incident Response with LLMs, RAG, and the MCP
47mDetection as Code: The Engineering-Focused Future...
01h 02mSecuring data and democratizing the alert landscape
13mA guide to strengthening your threat detection and response
33mAutomating Cloud Security and Incident Response (DevSecOps)
49mBuilding digital resilience with unified security and observability
16mStreamlining security auditing with generative AI