Mock interview practice with feedback after each answer

Act as an experienced interviewer for the role below. Run a mock interview with me.

Role: [JOB TITLE]
Company type: [e.g. "50-person fintech startup", "large public hospital"]
Interview stage: [e.g. "first-round screen", "final round with the team lead"]
Job ad or key requirements:
"""
[PASTE JOB AD OR REQUIREMENTS]
"""
My background in brief: [2-4 SENTENCES]

Rules:
- Ask exactly one question at a time and wait for my answer. Never answer for me.
- Mix question types: [e.g. "behavioural, role-specific, one about motivation, one curveball"].
- Ask a natural follow-up question when my answer is vague or missing a result.
- After each of my answers, give feedback in this format:
  Strength: one sentence.
  Improve: one or two specific suggestions.
  Score: 1-5 for clarity and 1-5 for relevance.
- After [NUMBER] questions, stop and give a summary: top 3 strengths, top 3 areas to rehearse, and 3 questions I should prepare for the real interview.

Start with a short greeting as the interviewer, then ask the first question.

Example (excerpt):
Interviewer: Thanks for joining. Tell me about a time a project you owned started to slip. What did you do?

(after the answer)
Strength: You named the cause of the delay clearly and took ownership.
Improve: Say what the outcome was. Did the project ship, and how late? Cut the background on the vendor; it took a third of your answer.
Score: clarity 3/5, relevance 4/5.

Follow-up: You mentioned re-planning with the team. How did you decide what to drop?

When to use it

Use this prompt in the days before an interview, when you know your stories but have not yet said them aloud under pressure. It works for screening calls, competency-based interviews and panel rounds. It is also useful if you are changing careers and need to practise explaining how your experience transfers.

How to customize it

• [JOB TITLE] and [COMPANY TYPE] shape the questions. A startup interviewer probes for ownership and speed; a large organization may focus on process and stakeholders.
• [INTERVIEW STAGE] sets the depth. A screen is broad; a final round goes deeper into your judgment.
• [JOB AD] lets the interviewer test the specific requirements, which is where real interviews usually go.
• [QUESTION TYPES] lets you focus practice on what scares you, such as salary or gaps in your CV.
• [NUMBER] controls length. Five to eight questions is a good session.

Why it works

The “one question at a time” rule turns the model into a turn-based conversation partner instead of a list generator. Without it, many assistants will produce ten questions with model answers, which feels useful but does not train you to answer on your feet.

The fixed feedback format keeps critique short and actionable, and the scores make progress visible across sessions. Follow-up questions are what separate a real interview from a quiz; asking the model to probe vague answers trains you to include outcomes and specifics. The role and context at the top give the model enough information to ask questions that sound like the real thing.

Tips and variations

• Record yourself answering the three summary questions and listen back; filler words and long pauses are easier to hear than to read.
• If your assistant has a voice mode, use it. Speaking your answers is much closer to the real interview than typing them.
• Add “Be a tough interviewer who pushes back” for a harder session once you feel comfortable.
• Paste your answer notes afterwards and ask: “Turn my best answers into STAR bullet notes I can review.”
• Treat the feedback as a second opinion, not a verdict; a human mentor who knows the industry is still worth asking.
• See our guide to preparing for an AI or ML job interview for role-specific advice.