The most useful ChatGPT prompts for interview preparation belong to a stage of prep: decoding the job, researching the employer, building answers from your real experience, rehearsing, reviewing and the last 24 hours. Below are 14 copy-paste prompts in that order. They are plain text, so they work the same in ChatGPT, Gemini, Claude and Microsoft Copilot.
This page is the prompt library. For the method behind it (you do the thinking first, then let the AI critique), read our guide on how to use AI for interview prep. For full mock-interview scripts by round, see our ChatGPT mock interview prompts; the prompts here cover everything around the mock.
Before you start: what to paste and what to strip out
Every prompt below assumes you paste three things: the full job description, your résumé (or the roles that matter), and your own notes. Before you paste, strip out:
- Your current employer's confidential information: unreleased products, client names, internal figures. Use placeholders such as [client A].
- Other people's names and personal details. Use roles instead: my manager, a client in retail.
- Your own phone number, home address and ID numbers.
Fill every [square bracket] before you send, and start a fresh chat for each stage so the instructions stay in focus. The stages map onto a week like this:
| Stage | When | Prompts | Done when |
|---|---|---|---|
| Decode the job | The day you get the invite | 1 to 3 | You can name the three problems the role exists to solve |
| Research and forecast | 3 to 7 days out | 4 and 5 | Three verified facts and a ranked list of likely questions |
| Build answers | 3 to 6 days out | 6 to 9 | Every likely question has a true story or point behind it |
| Rehearse and review | 2 to 5 days out | 10 to 12 | Two spoken rounds done, one fix chosen after each |
| Last 24 hours | The day before and the morning of | 13 and 14 | One page of notes and no new material |
If you want the same plan as a tick-box list with time budgets, our AI interview prep checklist lays it out by day.
Stage 1: decode the job
1. Read the job description as its author
You wrote this job description as the hiring manager. Separate the must-have requirements from the nice-to-haves, rewrite each must-have as the problem the new hire has to solve in their first six months, and point out which lines are boilerplate. Job description: [paste]
Interviews probe problems, not bullet points. The output tells you what your stories need to prove; check it against what the recruiter says about the role.
2. Gap map against your résumé
Compare my résumé with this job description. Make a table with three columns: requirement, my strongest evidence (quote the exact résumé line), and the doubt an interviewer might still have. If there is no evidence, write 'none'. Do not invent experience. Résumé: [paste] Job description: [paste]
The third column is your list of questions to prepare for. The 'none' rows show where you need a transferable example or an honest answer about how you'd learn it.
3. Map the process from the recruiter's email
Here is my recruiter's email describing the interview process, with names removed: [paste]. For each stage, tell me what it is most likely testing, who I'm likely to meet and what to prepare. Then list the questions I should ask the recruiter to fill the gaps. Label anything you're guessing as a guess.
It turns 'a few conversations with the team' into a plan, and the recruiter questions get you facts the model can only guess at.
Stage 2: research the employer and forecast the questions
4. Turn your research notes into talking points
Here are notes I took from [company]'s own website, newsroom and latest annual report or blog: [paste]. Using only these notes, give me three talking points that connect the company's current priorities to my background, which is: [two-line summary]. Add one question to ask about each. Flag anything in my notes that looks out of date.
'Using only these notes' stops the model filling gaps from memory, which may be stale. Our guide to researching a company before an interview with AI covers how to gather the notes.
5. Questions by interviewer type
For a [role] interview at [level], list the four questions each of these people would most likely ask me: the recruiter, the hiring manager, a future peer and a senior leader. For each question, say in one line what that person is really checking. Don't answer them.
One hiring process has several audiences, each with a different worry. The free interview questions from a job description tool gives you a second list to merge with this one.
Stage 3: build answers from your own experience
6. Pull a story out of rough notes
I'll paste rough notes about a project: [paste]. Interview me about it one question at a time to pull out what a strong interview story needs: what was at stake, the decisions I made myself, what got in the way, a result I can verify and what I'd change. After eight questions, summarize my answers as STAR bullet points in my own words. Add nothing I didn't say.
The model asks and you supply the facts, so the story stays true. To build a full set of stories, use the workflow in building a STAR story bank with AI.
7. Proof check
Label each claim in my answer below as proven (a number, a name or a decision I made), vague, or unsupported. For each vague claim, ask me one question that would help me add a real detail. Don't rewrite the answer. Answer: [paste]
Interviewers follow up on vague claims. Fixing them with your own details keeps the answer true and in your voice.
8. Turn a draft into a speaking outline
Here's my draft answer to '[question]': [paste]. Don't polish it. Turn it into a speaking outline of no more than five bullets of six words each, in the order I should say them, so I can practice from memory instead of reading.
A polished paragraph read aloud sounds recited; a short outline keeps every run natural.
9. Question-to-story matrix
Here are my stories, one line each: [paste]. Here are the questions I most expect: [paste]. Build a matrix showing which story can answer which question. Mark questions with no story, and flag any story I'm relying on for more than three questions.
It shows gaps and overused stories in one view, so you know where you need a new example.
Stages 4 and 5: rehearse and review
10. A mixed 25-minute round
Run a 25-minute practice interview for [role]: one motivation question, two questions about my experience, one about a gap or weak spot in my résumé, one situational question, then 'Do you have any questions for us?'. Ask one at a time and wait for my answer. Follow up only when my answer lacks a specific example. Save all feedback for the end.
Holding feedback until the end keeps the round closer to a real interview. Answer out loud even if you type afterwards; our ChatGPT voice mode guide covers spoken setups.
11. Uncomfortable questions for your background
Based on my résumé, ask me the three questions I'd least like to hear for this role, such as [a short tenure, a career change, a missing qualification], one at a time. After each answer, tell me in one line whether I sounded defensive and which phrase to drop.
These are the questions people skip in practice and stumble on in the room. The goal is a calm, honest answer, not a clever one.
12. Review a spoken transcript
Below is a transcript of my spoken answer to '[question]', from my phone's dictation or a recording. Count the filler words, give the word count, mark the sentence where I lost the thread, and tell me whether my first sentence answered the question. Transcript: [paste]
About 250 words takes roughly two minutes at a calm pace, a sensible ceiling for most experience questions. Pick one fix per session rather than five.
Stage 6: the last 24 hours
13. One-page prep sheet
From everything in this chat, build a one-page prep sheet: bullets for my 60-second introduction, five stories with a one-line hook each, the three numbers I must remember, three questions to ask, and blanks for the time, the call link and the interviewers' job titles. Keep every item to one line.
One page you can review in five minutes. Check every number on it against your own records.
14. Morning-of warm-up
Ask me eight quick questions from the list in this chat, in random order. I'll answer each out loud in under 45 seconds and type 'next'. Only tell me whether I opened with my main point. No other feedback today.
The morning of an interview is for confidence, not new material. One narrow check keeps you sharp without opening new doubts.
Prompts tuned to a specific round go deeper than this general set:
- behavioral interview prompts for stories, ownership and follow-ups
- technical interview prompts, coding interview prompts and system design prompts
- case interview prompts for consulting and strategy roles
- panel interview prompts, hiring manager prompts and final round prompts
- presentation interview prompts for briefs, slides and the Q&A
How to check what the AI gives you
A chatbot is fluent whether it is right or wrong, so check its output before any of it goes into your answers:
| Check | Why it matters | Quick test |
|---|---|---|
| Facts about the employer | Models can describe an old interview process or outdated figures with confidence | Confirm on the employer's own site or with your recruiter |
| Details you didn't give it | An added number or decision falls apart on the second follow-up | Ask: which details in this did I not give you? |
| Inflated claims | 'Led' when you contributed invites questions you can't answer | Keep only verbs you'd defend in conversation |
| Generic phrasing | 'Passionate team player' fits every candidate | Swap in a name, a number or a decision you made |
| Length | A written answer that looks short can take four minutes to say | Read it aloud with a timer |
| Generous scores | Chatbots lean positive, and scores drift between chats | Ask for the weakest sentence, not a grade |
The experience in every answer is yours; never memorize the AI's wording to recite.
Where CoPilot Interview fits
Prompts cover the thinking. What a text chat handles least well is spoken delivery with a bit of pressure. CoPilot Interview is an AI interview helper built for that part: its AI mock interview builds questions from your résumé and the job description, reads them aloud and scores each spoken answer out of 10. The free plan gives 15 minutes a day of live help in sessions of up to 15 minutes, with AI mock interviews included. Paid plans are Standard at $24.99 a month and Pro at $39.99 a month. It runs as a desktop app for Windows 10 and 11 and macOS 12 or later, a free iPhone and iPad companion on the App Store (iOS 16 or later) and an Android companion in beta. To see the scoring without installing anything, the free practice demo has 145 questions in 10 tracks.
It also offers live help, with captions of the interviewer's question and a structured answer in about two seconds. Use that only where it is allowed: follow your employer's rules, and treat online assessments, proctored tests and AI-led rounds as things to practice for, not places to use outside help.
Disclosure: CoPilot Interview is our product, so weigh what we say about it accordingly.
Frequently asked questions
What is the best ChatGPT prompt for interview preparation?
There isn't one; the best prompt depends on the stage. If you use only one, use the gap map (prompt 2). It compares your résumé with the job description and shows the doubts you need to answer.
Can I use ChatGPT to write my interview answers?
Use it to organize and critique answers you build from your own experience. Don't let it invent stories, and don't memorize its wording: interviewers follow up, and an answer you didn't live breaks on the second question.
Do these prompts work in Gemini, Claude and Microsoft Copilot?
Yes. They are plain text with no tool-specific features, so you can paste them into any general chatbot. Start a new chat for each stage.
How early should I start using AI to prepare for an interview?
A week is comfortable for most first and second rounds: decode the job and research early, build answers mid-week, and leave two or three days for spoken rehearsal. With only a day, run prompts 2, 10, 13 and 14.
Is it safe to paste my résumé into ChatGPT?
Usually, once you remove your phone number, address, other people's names and anything confidential about your employer. Review the privacy settings of the tool you use.
Try a practice question
The browser demo has 145 real interview questions across 10 tracks. Type your answer and get a score out of 10, strengths, areas to improve and a follow-up question. Free, no signup.
Try a practice question