Yueser ← Back to library

AI / Office & Communication

Competency-Based Interview Questions

One verifiable question per competency

Using the JD and level below, design a 45-minute interview. Start with four or five competencies this role actually needs, eight words or fewer each; give each one a main question plus two follow-ups, and make the main question ask what the candidate has actually done in the past. For every question, write what a passing and a failing answer sounds like. Leave five minutes for their questions. Do not ask brainteasers, horoscopes, or marital status.

[Paste JD]
[Level]
How to use

A good question forces a specific project. If the candidate can answer with “usually I would”, the question is not doing its job.

Good for

  • A new role draws a batch of resumes, three interviewers ask three different things, and nobody ends up testing whether the candidate can debug on their own.
  • You are hiring a senior person and the JD is all “familiar with” and “proficient in”; you need to turn those into questions that produce evidence in 45 minutes.

Not for

  • Campus or zero-experience candidates — focus on learning speed, a work sample, and coachability; asking what they did in a past job yields nothing.
  • You have not decided what problem the role solves — do the role design first, or the questions just rephrase the JD's adjectives.

Example

Input
JD: senior backend, owns payment-path reliability, familiar with high concurrency, has pushed work across teams. Level P6, 45-minute interview.
What you get
Start with four competencies: payment-path debugging, capacity planning, cross-team drive, and incident review. Give each a past-behavior question, such as “walk me through a payment timeout you diagnosed — how did you narrow it down, and what was the root cause?”, with follow-ups on the data, the alternatives, and what they would do differently. Set the bar — a pass names concrete metrics and trade-offs, a fail only names processes and tools — and leave five minutes for their questions.

Common mistakes

  • Asking “what would you do if...” — candidates recite a prepared answer, and you hear the rehearsal, not the work.
  • Trying to cover seven or eight competencies in 45 minutes — every one stays shallow, none is actually verified, and the interview only looks thorough.

Try tweaking it like this

  • Add “tag each question to the JD line it tests”, so after the loop you can see which competency nobody covered.
  • Add “two interviewers ask the same question separately and score before comparing”, for teams whose rounds do not line up.

More in this topic