Site Reliability Engineer Interview Question

How do you define toil, and how do you decide what to automate first?

What the interviewer is probing, how to structure your answer, and a spoken example you can adapt.

Quick answer

Toil is manual, repetitive, automatable work with no lasting value that scales linearly with the size of the service. Measure it before you fight it: log where on call and interrupt time goes for a few weeks. Then rank candidates by frequency times time cost times risk and automate the top one. Removing the need for a task entirely beats scripting it, and not all toil is worth automating.

Why interviewers ask this

This checks whether you can prioritize engineering effort with evidence rather than by whatever annoyed you most last week. Interviewers listen for measurement, for the distinction between automating a task and eliminating it, and for the honest recognition that a script which runs twice a year and needs maintenance can cost more than the toil it replaced.

How to structure your answer

  • Give the definition with the scaling property included.
  • Insist on measuring where the time actually goes first.
  • Rank by frequency, time cost, and risk of doing it manually.
  • Prefer eliminating the task over automating it.
  • Acknowledge when automation is not worth the maintenance.

Example answer

Spoken example, first person

Toil is the manual work that grows with your fleet and leaves nothing behind: rerunning failed jobs, granting access by hand, restarting the same service every Monday. The trap is automating whatever irritated you most recently, so I start by measuring. We tagged every interrupt and on call action for a month, and the answer was not what anyone predicted: a third of it was one team asking for the same database read only credential, which we replaced with a self service request in about two days of work. That is the second point, elimination beats automation. Nobody needed a script to grant credentials faster, they needed to not ask us at all. I also keep a floor on it. If a task takes ten minutes twice a year, a script for it is a liability, because it will be broken and untested the next time you need it. I aim for on call time under about half interrupt work, and track it as a real metric.

Walking into this interview soon? GhostPilot listens to your live call, spots the question the moment it is asked, and puts a structured answer on your screen in real time. Try it on your next mock, or grab a $29 Session Pass, no subscription, for the real thing.

See how it works

Follow-up questions to expect

  • How do you measure toil without adding more toil to measure it?
  • What would you do if the team spends 70% of its time on toil?
  • How do you get product leadership to fund reliability work?

Related site reliability engineer questions

Your interviewer will ask their own version of this. Paste your actual job description into the free Question Predictor and get the 20 questions that role is most likely to ask, with what each one is really probing.

Predict my questions

Rehearse the hard questions before they are asked

Practise with a live copilot, then walk in ready. A $29 Session Pass gets you through the interview with no subscription and no lock-in.

Get GhostPilot