Reliability Engineering for Humans
recording made private on YouTubehttps://youtu.be/aP5QB1kcykM
summary
Explains how concrete SRE practices map to human outcomes: define SLIs and SLOs, convert a 99% SLO into an 403 minutes per 28 days error budget, and use that budget to reprioritize backlog. Recommends blackbox SLIs, synthetic monitoring, and automation patterns like automated detection and recovery and automated rollback / blue‑green deploys. Shares an outcome: adopting Google SRE practices plus PagerDuty-derived alerts and health metrics moves a team’s health score from 20s to 90s and reduces MTTR.
AI-written summary · about
related talks
27mGetting Started with SRE
30mImproving Reliability w/ Error Budgets & SRE
22mSRE for Enterprises
38mImplementing SRE practices: SLI/SLO deep dive
30mReliability Engineering for Humans
47mSRE, Myths vs Realities
25mUnleashing the potential of SRE for modern operations
31mSite Reliability Engineering (SRE) and...