DevOps Engineer interview questions (India, 2026): what to ask and what to listen for
12 interview questions for a DevOps Engineer role, grouped by what each round should establish, each with what a strong answer contains, plus a five-point scorecard so every interviewer rates the same things.
Last updated 3 September 2026. Competitor pricing and plan limits are re-verified against each vendor’s own pages.
What they have run
- Describe the production setup you last ran: services, environments, pipeline, monitoring, backups, and the monthly bill as an order of magnitude. What would you simplify?
- A strong answer: A coherent picture with cost awareness and the judgement to name over-engineering. Candidates who cannot say what it cost have not owned it.
- Tell me about the worst incident you handled. What was the cause, how did you find it, and what changed afterwards?
- A strong answer: Composure, a diagnostic method, honest root cause, and a durable fix — alerting, a runbook, a design change.
- When did you last restore from backup, and what did you learn?
- A strong answer: A real restore or drill with a discovered gap — wrong retention, missing secrets, a slow restore. A backup never restored is not a backup.
- Describe a cost reduction you drove. What did you find and what did you change?
- A strong answer: Specific waste — idle instances, oversized databases, egress, logs retention — and a measured saving.
Technical depth
- Design the infrastructure for [our product at our scale] from scratch, with a bill an SME can afford. Think aloud and justify each choice.
- A strong answer: Proportionate choices — managed services, a simple deploy model, monitoring that matters — with cost reasoning, not a platform-team blueprint.
- A deploy has broken production. Walk me through the next fifteen minutes.
- A strong answer: Rollback first, communicate, then diagnose; and a pipeline that makes rollback trivial.
- How do you manage secrets and access so that a leaked laptop or a departing engineer is not a disaster?
- A strong answer: A secrets manager, no secrets in code or CI logs, least-privilege IAM, rotation, offboarding checklist.
- Alerts are firing all night and most are noise. What do you do?
- A strong answer: Tunes on symptoms not causes, deletes alerts nobody acts on, adds severity, and reviews after incidents — treats alert quality as engineering work.
Working with a small team
- A developer wants to add a new managed service that costs money. How do you evaluate it?
- A strong answer: Understands the need, weighs cost against time saved and operational load, proposes alternatives, and does not default to no.
- How do you write a runbook someone else can follow at 3 am?
- A strong answer: Steps, commands, expected output, what to do if it does not work, who to call — tested by someone other than the author.
- A customer sends a security questionnaire. What can you answer today and what would you need to build?
- A strong answer: Familiarity with the usual questions — encryption, access, backups, incident response — and honesty about gaps.
- What do you want to get deeper in — security, data, platform, a cloud — and does this role offer it?
- A strong answer: A direction that fits a sole-DevOps role — security posture, cost engineering, database operations — rather than one that needs a platform team. Candidates whose ambition is large-scale Kubernetes platforms will find an SME setup too small within a year.
Scorecard for DevOps Engineer interviews
Score each candidate on these five criteria, one to four, straight after the interview and before talking to the other interviewers. Written scores taken independently are what make a panel decision defensible; a discussion first produces one opinion with several signatures.
- Production ownership: ran a real setup and knows what it cost.
- Incident and recovery maturity: incidents handled, restores performed.
- Proportionate design: chooses simple, affordable, reliable options with reasons.
- Security and access discipline: secrets, IAM, offboarding, questionnaires.
- Team enablement: runbooks, developer support, cost conversations.
Screen before you interview
Most DevOps Engineer interviews that go badly were avoidable at the application stage. The job description template for this role carries five screening questions that settle the deal-breakers — notice period, location, the one or two hard requirements — before anyone books a slot.
- How many years have you run production infrastructure on [cloud provider] for a product with real users? — knockout
- Have you personally performed a backup restore in production or a realistic drill? — for context
- Are you able to work [from our City office / remote with overlap hours] and take part in an escalation rota? — knockout
- What is your current notice period, in days? — for context
Questions, answered straight
What is the best DevOps Engineer interview question?
Ask when they last restored from backup and what they learned. It is the single most telling question about operational maturity, and candidates who have done it always have a story about a gap they found.
Should a DevOps interview include a design exercise?
Yes, sized to your product: design the infrastructure for it from scratch with a bill an SME can afford, thought aloud. Proportion and cost reasoning are what you are assessing, not knowledge of every service.
How do I assess cost awareness?
Ask what their last production setup cost per month as an order of magnitude and what they did to reduce a bill. Engineers who have owned the bill answer with numbers and specific waste they removed.