GhostpilotHome · Interview questions · Site Reliability Engineer
Site Reliability Engineer interview questions
SRE interviews combine systems knowledge with software thinking. You'll be asked to reason about failure, measure reliability and explain how you'd stop the same incident happening twice.
What interviewers look for
- Systems depth. Linux, networking and distributed systems.
- Reliability thinking. SLOs, error budgets and blameless reviews.
- Automation. Replacing repeated manual work with code.
How the interview usually runs
- Recruiter call.
- Coding or scripting screen.
- Troubleshooting interview.
- System design for reliability, then a behavioural round.
Stages vary by company, so ask the recruiter what to expect before you start preparing.
10 Site Reliability Engineer interview questions, and how to answer them
1. What's the difference between an SLI, SLO and SLA?
What they want to see: Core vocabulary.
How to answer: A measurement, a target, and a contract with consequences. Give an example for an API.
2. How do you use an error budget?
What they want to see: Balancing speed and safety.
How to answer: When it's spent, slow releases and spend on reliability, by agreement with product.
3. A service's latency doubled. Walk me through it.
What they want to see: Live troubleshooting.
How to answer: Check recent changes, saturation and dependencies, then use traces to find the slow part.
4. How do you run a postmortem?
What they want to see: Learning culture.
How to answer: Keep it blameless, with a timeline, contributing causes and owned actions.
5. What is toil and how do you reduce it?
What they want to see: Engineering mindset.
How to answer: Repetitive manual work that scales with load. Measure it and automate the worst first.
6. How would you make a service multi-region?
What they want to see: Distributed design.
How to answer: Discuss traffic routing, data replication, consistency and failover tests.
7. What happens during a TCP handshake?
What they want to see: Networking fundamentals.
How to answer: SYN, SYN-ACK, ACK. Mention how it shows up when debugging timeouts.
8. How do you plan capacity?
What they want to see: Forecasting.
How to answer: Use growth trends and load tests, and keep headroom for peaks and failures.
9. What makes a good alert?
What they want to see: Avoiding alert fatigue.
How to answer: It's urgent, actionable and tied to user impact.
10. Tell me about a reliability improvement you drove.
What they want to see: Impact.
How to answer: Give the before and after, in availability or incident count.
A sample answer
How do you use an error budget?
Our checkout SLO was 99.9%. After a bad month we'd used the budget, so we agreed with product to pause feature releases for two weeks. We fixed the two top causes: connection pool exhaustion and a flaky dependency. The next quarter stayed inside budget, and product now joins the monthly review.
A made-up example to show the structure. Use your own experience and real numbers.Common mistakes
- Giving definitions without examples.
- Treating every incident as someone's fault.
- Alerting on causes rather than symptoms.
- Having no numbers for reliability work.
Practise this out loud
Run a mock interview for a Site Reliability Engineer role. Questions change every time, you answer by voice or typing, and each answer gets a score with feedback.
Frequently asked questions
How is SRE different from DevOps?
They overlap. SRE puts more weight on measured reliability and software engineering, while DevOps roles often focus more on delivery pipelines.
How much coding do SRE interviews include?
Usually more than DevOps roles: expect practical coding in Python or Go, and sometimes algorithms.