DevOps & SRE Interview Question Bank

I sat through 78 interviews across 43 companies in two and a half months. Three of them ended in offers.

By interview 20, the pattern was obvious. The tool questions were never the hard part. The ones that decided the outcome were the judgement questions — what did you check first, what did you get wrong, what would you do at 3am with no runbook.

So I wrote all of it down.

What's inside — 24 pages, 13 sections:

• Linux — the section everyone underestimates

• Networking

• Git & branching

• Docker & containers

• Kubernetes — the longest section

• CI/CD pipelines

• Terraform & Infrastructure as Code

• Cloud — AWS and Azure

• Monitoring & observability

• Scripting — Bash and Python

• Scenario & incident rounds

• SRE practice — SLOs, on-call, postmortems

• How to actually use this

What makes it different: these aren't definitions copied from documentation. Every answer is written the way you'd actually say it in the room, and most of them include what happens when the thing is configured wrong — which is the difference between a 2-year answer and a 5-year answer.

Written for 1–6 years of experience. Free.

— Om Jaju, Site Reliability Engineer, ex-Morgan Stanley