What SRE Means in a Start-up
summary
Describes how SRE at Intercom runs on AWS with ELB, nginx and a monolithic Ruby on Rails app, with about 60 engineers in Dublin, owning the build and deploy pipeline, paging via PagerDuty and metrics from Datadog/New Relic/logentries. Argues startups should run less software: avoid heavyweight change control, prefer lightweight changes, and buy SaaS instead of reimplementing. Shows team rotation across product and infrastructure, a limit to how many SRE teams scale without moving from monolith to microservices, and practical tradeoffs backed by concrete tooling and org counts.
AI-written summary · about