ChatGPT prompts for job interview prep

These 14 ChatGPT prompts for job interview prep help you decode the role, turn your real experience into STAR answers, rehearse in a mock interview and choose good questions to ask at the end. They work the same way in Claude and Gemini. Every prompt asks for your own material, because an answer built from your real work is the only kind that survives a follow-up question.

How to use these prompts

Copy a prompt, replace the blanks in double braces, or keep the default after the colon: {{role:product manager}} means "product manager" unless you change it. Blanks with no default, like {{job_description}}, need your own text.

Gather three things before you start:

• The job description, pasted in full, including the "nice to have" list.
• Your raw notes: projects, results, mistakes and what you learned, in plain bullet points. Rough is fine.
• What you know about the interview: the format, who you will meet, and how long it lasts.

OpenAI's prompting guide for ChatGPT at work suggests treating a prompt like a handoff to a new intern: give the background, the task and what good looks like. That advice matters more for interview prep than almost anything else, because a model with no context will invent a generic candidate.

Two rules keep the output honest. First, tell the model to use only what is in your notes and to ask you when something is missing. Anthropic's guide to reducing hallucinations recommends a related step: explicitly give the model permission to admit uncertainty. Second, leave confidential details out. Describe clients and numbers in general terms, the same way you would in the room.

If you still need to fix your CV, start with the resume prompts. For coding rounds and system design, use the technical interview prompts.

Which prompts help you understand the role?

1. Decode the job description

Job ads mix must-haves, wishes and boilerplate. This sorts them and shows where your gaps are before the interviewer does.

Here is a job description for a {{role:product manager}} role and my notes
on my experience.

1. List the 5 skills or experiences this employer cares about most, ranked,
   with the line from the ad that tells you so.
2. For each one, find the strongest matching evidence in my notes. If there
   is none, write "GAP" and suggest how I could honestly address it.
3. Name 3 questions I am likely to be asked because of the gaps.

Use only my notes as evidence. Do not invent experience.

Job description:
{{job_description}}

My notes:
{{my_notes}}

2. Company research brief

Paste the company's own pages (about page, recent announcements, product pages) rather than asking the model to remember the company, which it may get wrong or out of date.

I have an interview at {{company:the company}} for a {{role:product manager}}
role. Below is text I copied from their website and recent announcements.

From this text only, give me:
- What they sell and to whom, in two sentences
- Their stated priorities or recent changes
- 3 ways my background connects to those priorities (use my notes)
- 3 things I should find out in the interview

If the text doesn't answer something, say so instead of guessing.

Company text:
{{company_text}}

My notes:
{{my_notes}}

3. "Tell me about yourself"

Almost every interview opens with it. A clear 60 to 90 second answer sets the tone for the rest.

Write three versions of my answer to "Tell me about yourself" for a
{{role:product manager}} interview. Each should take about
{{length:75 seconds}} to say out loud.

Structure: where I am now, the one or two experiences that led here, and
why this role is the logical next step. End on the role, not my history.
Plain spoken language, no buzzwords. Use only facts from my notes.

My notes:
{{my_notes}}

How do you build STAR answers with ChatGPT?

STAR (Situation, Task, Action, Result) is the usual shape for behavioural answers. Most of the time should go on the Action, and the Result should say what changed. These prompts start from your stories, not from a list of perfect answers.

4. Build a story bank

Six to eight good stories cover most behavioural questions. This finds them in your notes and maps them to common themes.

From my notes below, pick 6 to 8 stories I could use in behavioural
interviews. For each story give:
- A short title
- The themes it can answer (leadership, conflict, failure, influence,
  ambiguity, deadline pressure, customer focus, learning something fast)
- What is missing that I should remember before the interview, such as a
  number, a date or what happened next

Then show a table of themes against stories so I can see which themes have
no story yet.

My notes:
{{my_notes}}

5. Turn one story into a STAR answer

Give it the raw version, the way you would tell a friend. Asking for questions back catches the details you forgot.

Turn this story into a STAR answer for the question
"{{question:Tell me about a time you handled a difficult stakeholder.}}"

Situation and Task: 2 or 3 sentences together.
Action: most of the answer, in first person, with what I did and why.
Result: what changed, with numbers only if they are in my story, plus one
line on what I learned.

Keep it under {{length:2 minutes}} spoken. Before writing, ask me up to 3
questions if anything important is missing. Do not invent results.

My story:
{{story}}

6. Critique an answer you drafted

For an answer you already wrote or recorded. A harsh but specific reviewer beats a polite one.

Act as an experienced interviewer for a {{role:product manager}} role. Here
is the question and my answer.

Score it 1 to 5 on: relevance to the question, clarity of my own actions,
evidence of results, and length. For each score under 4, quote the part
that caused it and show a better version of just that part. Then list the
follow-up question you would ask to test whether the story is real.

Question: {{question}}
My answer: {{answer}}

7. Weakness and failure questions

These are where generic AI answers sound most fake. This prompt pushes for a real weakness and real evidence of improvement.

Help me answer "{{question:What is your biggest weakness?}}" honestly.

Here is a real weakness or failure of mine and what I have done about it:
{{weakness}}

Write an answer that names the weakness plainly, gives one concrete example,
and shows what I changed and how I know it is working. No disguised
strengths like "I work too hard". Under 90 seconds spoken. If my notes show
no real improvement yet, tell me what I could start doing this week.

Which prompts prepare you for role-specific questions?

8. Likely questions for this role

A ranked list you can practice against, split by type so you don't over-prepare one kind.

List the {{count:15}} questions I am most likely to be asked in a
{{stage:first-round}} interview for a {{role:product manager}} role at a
{{company_type:B2B software company}}, based on the job description below.

Group them as: behavioural, role skills, situational ("what would you do
if"), and motivation. For each, add one line on what the interviewer is
really checking. Mark the 5 I should prepare first.

Job description:
{{job_description}}

9. Situational and case questions

The US Office of Personnel Management describes a structured interview as asking candidates about "their behavior in past experiences and/or their proposed behavior in hypothetical situations." This prompt rehearses the second kind.

Give me {{count:3}} realistic "what would you do if" scenarios a
{{role:product manager}} might face in their first 90 days at a
{{company_type:B2B software company}}.

For each, wait for my answer before continuing. Then tell me what a strong
answer would cover (questions to ask first, trade-offs, who to involve),
and where my answer was vague or skipped a step. Start with scenario 1 only.

10. Explain a gap or a career change

Short, factual and forward-looking is the goal. The prompt stops the answer from turning into an apology.

Help me explain {{situation:a 14-month career gap}} in an interview for a
{{role:product manager}} role.

What actually happened: {{facts}}
What I did during that time that is relevant: {{relevant}}

Write a 30 to 45 second answer that states the facts briefly, links to
something useful for this role, and moves on. No over-explaining and no
apology. Then give a one-line answer for a follow-up like "Why now?".

How do you run a mock interview with AI?

Typing answers is useful, but speaking them is closer to the real thing. Claude's voice mode is a beta feature that Anthropic says is "built to work best from your phone", and it has a push-to-talk option where you hold a button while speaking and release it when you're done, which Anthropic says gives you more control over when Claude listens.

11. Full mock interview, one question at a time

The instruction to wait is the important part. Without it, the model asks and answers its own questions.

Run a {{length:30-minute}} mock {{stage:first-round}} interview for a
{{role:product manager}} role. You are the interviewer.

Rules:
- Ask one question at a time and wait for my answer.
- Ask a follow-up when my answer is vague, the way a real interviewer would.
- Mix behavioural, role and motivation questions, based on the job
  description below.
- Do not give feedback until I type "end interview".

When I type "end interview", give me: my 3 strongest answers, my 3 weakest
with a better outline for each, and any habits to fix (rambling, "we"
instead of "I", no results).

Job description:
{{job_description}}

12. Tough interviewer and scorecard

For the round you are most nervous about. A scorecard turns "it felt bad" into specific things to practice.

Act as a skeptical {{interviewer:hiring manager}} interviewing me for a
{{role:product manager}} role. Push back on my answers, ask for specifics
and numbers, and challenge weak reasoning, but stay professional.

Ask {{count:6}} questions, one at a time, waiting for each answer.

At the end, fill in this scorecard for each competency in the job
description: rating 1 to 4, the evidence I gave, and what was missing. Then
tell me the one change that would most improve my next interview.

Job description:
{{job_description}}

What should you ask the interviewer?

13. Questions to ask at the end

Good questions show you thought about the job, and they help you decide if you want it. Generic ones waste the slot.

Suggest {{count:10}} questions I could ask at the end of a
{{stage:first-round}} interview for a {{role:product manager}} role, with
{{interviewer:the hiring manager}}.

Base them on the job description and company text below. Group them as:
the work itself, how success is measured, the team, and growth. Skip
anything the text already answers. Mark the 3 that would tell me most about
whether I want this job.

Job description:
{{job_description}}

Company text:
{{company_text}}

14. Thank-you note after the interview

Short and specific. Mention one thing from the conversation so it doesn't read like a template.

Write a short thank-you email to {{name:the interviewer}} after my
{{stage:first-round}} interview for the {{role:product manager}} role.

One thing we discussed that I want to mention: {{detail}}
Anything I want to add or correct from the interview: {{addition:nothing}}

Under 120 words. Warm but not gushing. No repeating my whole pitch.
Include a subject line.

What should you not let AI do in interview prep?

AI is good at structure, practice and feedback. It should not supply the substance:

Never let a model add results, titles or tools you haven't used. Interviewers ask follow-up questions, and invented details fall apart under the second one. The same goes for reference checks and background checks later in the process.

Mind what you paste. On consumer Claude plans, Anthropic's model training policy explains when chats can be used to improve Claude and says Incognito chats are not used for that. Whatever tool you use, keep client names and internal figures out of it.

For more on how interviewers build structured questions, the recruiting prompts show the other side of the table. For the follow-up email if you don't hear back, see the email writing prompts.

How do you keep your prep prompts ready for the next interview?

Most job searches involve several interviews in a few weeks, and you will reuse the same prompts each time with a new job description. Kipt keeps prompts like these as Markdown in a private GitHub repo you own: save the mock interview and STAR prompts once, tag them "interview", find them with Cmd+K, and fill in the blanks for each new role. If you would rather run one as an Agent Skill, paste it into the free SKILL.md generator.