AI Teammates: How will they impact SRE Teams?
summary
Newberg’s Hawkeye claims to automate SRE root‑cause work by connecting to CloudWatch, Prometheus, GitHub, Jenkins X, DynamoDB and PagerDuty, virtualizing telemetry so LLMs write code in real time to query logs and configs and then execute that code. In the EKS workshop demo it correlates a Jenkins X commit that flipped a DynamoDB create flag, produced a deterministic RCA and remediation steps, and restores service — no automatic fixes today but integrated alerts, RCAs and follow‑ups reduce investigation time. agentic AI data access virtualization
AI-written summary · about
chapters (12)
- 0:00Introduction to AI teammates
- 1:04SRE toil and AI need
- 3:15Limitations of AIOps
- 5:01Observability tools today
- 6:49Agentic AI for SREs
- 8:54Demo scenario setup
- 13:09Hawkeye investigation demo
- 17:24Automated RCA generation
- 20:40Integrations and security
- 22:46Accuracy and usefulness
- 24:57Workflow integration
- 26:01Future of SRE AI tools
related talks
40mWhen Incidents Fix Themselves: AI SRE in action
26mAI & SRE: Embracing the Ultimate Frontier of Automation
23mBeyond Single-Cloud SRE
16mHow AI is redefining SRE and Customer Experience
01h 00mData is out, knowing is in: Applying AI to automate cloud operations
44mFrom 4 Hours to 8 Minutes with AI Agents That Transform SRE...
28mAI for Everything After Code
30mAI Agents for Incident Management