Site Reliability Engineer

About this role

Infrastructure owns the path from commit to production and everything that keeps it standing: build, deploy, observability, capacity and incident response.

Own availability across the platform: SLOs, on-call, capacity, and the postmortems that stop the same incident happening twice.

Multi-region, containerised, with tenant isolation that has to hold under load. The interesting problems are in resource isolation and in making a deploy boring.

Primary responsibilities

• Own the reliability of the production platform — define the SLOs, measure them, and use the error budget as a real input to what gets built
• Lead incident response as incident commander, and write the blameless postmortem afterwards
• Engineer away the toil: automate the recurring manual work rather than staffing it
• Build capacity models and plan ahead of demand rather than after an outage
• Improve the alerting so people are paged for things that matter and not for things that do not
• Run game days and failure injection to find out what actually happens rather than what is assumed

Skills required

Day to day this role works with Kubernetes, Observability, Incident Response and Go.

• Five or more years in SRE or a comparable operations role
• Have led an incident and written the review afterwards
• Four or more years in SRE, infrastructure or backend engineering with production ownership
• SLOs and error budgets used as real inputs rather than as vocabulary
• Incident command experience and blameless postmortem writing
• Strong coding ability — automating toil is most of the job