Beyond Single-Cloud SRE
summary
Demonstration shows an Azure SRE Agent and Hawkeye (Newbird) collaborating across an Azure frontend and AWS backend: a ServiceNow incident triggers parallel investigations, Hawkeye identifies an AWS API Gateway fault while Azure finds an Azure resource error, and the agents post a combined RCA and automated mitigation back into the ServiceNow discussion. Agents communicate over an API, use LLM hypothesis scoring, run LLM-based intent tests and golden end-to-end suites, and persist context in ServiceNow shared memory and GitHub for repeatable fixes.
AI-written summary · about
chapters (9)
- 0:00Introduction and Welcome
- 0:21Challenges in Multi-Cloud SRE
- 2:09Vision for AI Agents Collaboration
- 3:19Azure SRE Agent Overview
- 5:04New Bird SRE Agent Introduction
- 8:07Live Demo of Agent Collaboration
- 12:44Technical Insights and Architecture
- 17:59Shared Memory and Context Retention
- 21:53Declarative Agent Configuration
related talks
27mAI Teammates: How will they impact SRE Teams?
40mWhen Incidents Fix Themselves: AI SRE in action
24mAI SRE and Agent Handoffs: incident.io's Collaboration Model
36mSRE'ing the SRE AI Agents
44mFrom 4 Hours to 8 Minutes with AI Agents That Transform SRE...
34mAutomating SRE Operations with Multi-Agent AI
46mAI Agents for the Rest of Us
13mManaging Multi-Cloud Complexity