Why split questions into categories?

Behavioral, technical, and situational questions probe different signals — past performance, current ability, and judgment under hypotheticals. A structured mix gives a fuller, fairer read than asking all of one type.

Does it help avoid biased questions?

Yes. The generated prompt explicitly instructs the model to avoid illegal or discriminatory topics such as age, family status, religion, and health, and to keep every question job-relevant and neutrally worded.

What are the "what to listen for" notes?

Each generated question comes with a one-line description of what a strong answer looks like, so interviewers across a panel score consistently rather than relying on gut feel.

How many questions should I generate?

For a single round, eight to twelve is usually plenty. Generate more if you want a bank to draw from, then pick the subset that best maps to the competencies that matter most for the role.

Interview Question Prompt Builder

Interview question prompt builder

Good interviews are structured: the same competencies, probed consistently, with a shared idea of what a strong answer sounds like. This tool builds an LLM prompt that generates exactly that — a calibrated, categorized set of interview questions for a specific role and seniority, complete with answer-listening notes and bias-avoidance rules. It runs locally in your browser.

How it works

You enter the job role, pick the seniority, and list the competencies you want to probe. You choose an interview format and a question count. The builder produces a prompt that instructs the model to split questions across behavioral, technical, and situational categories, calibrate difficulty to the seniority level, map each question to a competency so nothing is missed, attach a one-line “what to listen for” note to each, and avoid illegal or discriminatory topics.

Why structured question banks matter

Unstructured interviews — where each interviewer asks whatever feels relevant — produce three consistent failure modes. First, the same candidate gets very different experiences depending on who interviews them, making panel scores incomparable. Second, certain competencies get over-probed (usually the ones the interviewer personally values) while others are ignored. Third, without explicit “what to listen for” guidance, interviewers often default to pattern-matching on background rather than measuring the actual competency.

A structured bank with coverage guarantees and scoring notes addresses all three. The prompt this tool generates instructs the model to maintain explicit competency coverage, calibrate difficulty to seniority, and attach a listening note so every panellist is scoring against the same standard.

Question types and when to use each

Behavioural questions (“Tell me about a time when…”) are the most predictive for experienced candidates because they anchor answers in past real events. Candidates who have genuinely done the thing tell a specific story with a specific outcome; candidates who have not tend to give hypothetical or general answers.

Technical questions probe current ability and are most valuable for roles where the specific skill is verifiable (coding, statistical analysis, financial modelling). For senior roles, deep technical questions matter less than judgment questions about architecture, tradeoffs, and failure modes.

Situational questions (“What would you do if…”) are most useful for roles without directly parallel past experience — for example, a first manager, a new graduate, or someone making a significant domain switch. They reveal problem-solving approach when historical evidence is limited.

Worked example

For a senior data analyst role, competencies might be: analytical rigour, stakeholder communication, SQL proficiency, and data integrity judgment.

The generated prompt would produce, for example:

Behavioural / analytical rigour: “Tell me about a time an analysis you delivered was later found to be incorrect. What happened and what changed?” Listen for: how they detected the error, whether they proactively communicated it, and what process change they implemented.
Situational / stakeholder communication: “A VP is asking for a report that you believe answers the wrong question. How do you handle that?” Listen for: confidence in challenging up, quality of alternative framing, whether they push back or just comply.

Tips and notes

List the competencies that genuinely differentiate strong hires for this role rather than a generic wish-list — the prompt guarantees each one is probed, so a bloated list dilutes the interview. For senior roles, the prompt deliberately steers away from trivia toward judgment and trade-off questions. After generating, review the bias-avoidance results yourself; the model is instructed to stay job-relevant, but you remain responsible for compliance with local employment law. Keep the “what to listen for” notes in your scorecard so a panel rates answers against the same bar.