Write interview questions for a role and level
Turn a role and seniority level into behavioural and technical interview questions, each paired with a note on what a good answer looks like.
Last updated
Fill in the blanks
Assembled in your browser — nothing you type is stored or sent anywhere.
Your prompt
Model-agnostic — works in Claude, ChatGPT, Gemini or any AI chat.
Why this prompt works
- →Requiring a "what a good answer looks like" note per question is what makes the output usable by an interviewer who isn't already an expert in the area being probed, not just a list of things to ask.
- →Calibrating the good-answer note to the stated level, rather than a single generic standard, prevents the common mistake of using the same rubric to evaluate a junior and a senior candidate.
- →Limiting technical questions to the given focus areas, with a prompt to flag ones that are too broad to question fairly, stops the model from inventing a plausible-sounding but ungrounded deep-dive question about a topic nobody asked for.
Part by part
- 4-5 behavioural questions and 4-5 technical or role-specific questions
- A balance rather than a count. Interviews that drift entirely technical or entirely behavioural test half the job.
- a short note on what a good answer looks like at this level
- The part that makes the questions usable by someone else. A question without a rubric is evaluated on impression, which is where bias concentrates.
- an interviewer with less experience in this area could still evaluate the response
- Names who the rubric is for. It is what turns a personal question list into something a panel can run consistently.
- a good answer from a senior candidate should look different
- Forces level calibration. The same question asked at two levels needs two different bars, and unstated bars default to the interviewer's own.
- say so and suggest what would need to be narrowed down
- Guards against unfair questions. A question on a focus area too broad to answer well tests confidence rather than competence.
Common mistakes
Write interview questions for a senior developer.Produces a generic list findable in any interview guide, which means candidates have rehearsed them. Without focus areas nothing is tested that relates to this particular job.
Write hard interview questions.Difficulty is not signal. A hard question with no rubric selects for people who interview well, and puzzle questions in particular have been repeatedly shown not to predict performance.
Write questions to test if they're a good culture fit.Culture fit is the least well-defined criterion in hiring and the one most prone to bias, because it rewards similarity. Ask about specific working practices instead — how they handle disagreement, or feedback.
Questions
Why does every question need a rubric?
Because without one, evaluation happens on impression, and impression is where bias operates. A written standard also lets several interviewers compare candidates rather than comparing their own reactions.
How should the level change the answer?
Usually in scope and in what is volunteered. A junior answer describes what they did; a senior answer describes the trade-off they weighed and what they would do differently — and the rubric should say so.
How many questions fit an hour?
Four or five with genuine follow-up. Nine questions in an hour means six minutes each, which produces rehearsed surface answers — the follow-up is where the actual signal is.
Should every candidate get the same questions?
Yes, for the core set. Structured interviews predict performance substantially better than unstructured ones, and consistency is also what makes comparing two candidates a fair exercise rather than a memory test.