At a glance
The interview-prep tree is shorter than people think. Behavioral round? Use prompts 1-3. Technical screen? Use 4. Case round? Use 5. Reverse-interview the recruiter at the end? Use 6. Asked the question that catches everyone ("why this company, specifically?") Use 7. Each prompt is short. Each one ends with you having something to say in your own voice, not a script you read.
Which prompt do you need for which interview round?
|
+– BEHAVIORAL (most rounds 2-4) —-> prompts 1, 2, 3
|
+– TECHNICAL SCREEN —————> prompt 4
|
+– CASE / WORK-SAMPLE ————-> prompt 5
|
+– "DO YOU HAVE ANY QUESTIONS?" —> prompt 6
|
+– THE SURPRISE QUESTION ———-> prompt 7
Interview prep usually fails one of two ways. Either it is too generic (rehearsing “tell me about yourself” without naming the actual company or role), or it is too rigid (memorizing answers that come out flat). The seven prompts below build muscle memory in a different way: each prompt produces practice material in your voice, with specifics that hold up under follow-up questions, and with the awareness that the interviewer is allowed to ask “and then what?”
⚠️ AI rehearsal trap to avoid
AI will happily generate polished, sound-good answers that you cannot actually defend when the interviewer follows up. The prompts below all force you to supply the specifics yourself. AI helps with structure, framing, and pressure-testing; the substance has to be true. The pre-publish editing prompt for written work applies to spoken answers too: How to Edit AI Out of Your Writing.
What are the seven interview prompts?
1. Build your story bank
Here is the job description: [paste].
Help me build a story bank. For each of the 6 most common behavioral questions, identify:
– Which of my past experiences best answers it (cite the specific resume line)
– The STAR-structure breakdown (Situation, Task, Action, Result) in 4-5 sentences total
– The likely follow-up question the interviewer would ask
– The specific number, name, or detail that anchors the story (this is what I should remember even if I forget the framing)
The 6 questions: tell me about yourself, a time you led, a time you failed, a time you disagreed with a peer, your biggest accomplishment, why this role / why now.
Why this works: A story bank is portable. Once you have six solid stories with anchor details, you can route any behavioral question through one of them. The “anchor detail” output is what saves you when nerves erase your memory mid-sentence; you remember the number or the name, and the rest comes back. The “likely follow-up” output is the part that prevents the unprepared “and then what?” stall.
2. Pressure-test a story
You are a senior interviewer at [company]. Push on this story. Ask me the 5 follow-up questions that would expose weak spots:
– The metric I cited (how was it measured? was I the cause?)
– The decision I made (what alternatives did I reject? was my call objectively right?)
– The team I led (what did individual contributors think? would they say I led well?)
– The conflict I navigated (would the other party tell this story the same way?)
– The outcome I claimed (is the company / team better off in a way I can verify?)
For each question, draft the best candid answer I could give. If the answer would expose a real weakness, name it; do not invent a better outcome.
Why this works: Most interview-prep skips the follow-up rehearsal because it is uncomfortable. The follow-ups are where the interview is actually won or lost; the polished primary answer is forgotten by the next round, but the way you handled the pressure-test answer is remembered. Practicing the worst follow-up question on each story is the work that compounds.
3. Translate your story to this company’s values
Here is the company's public values / culture page: [paste].
Here is the role description with any "what we look for" section: [paste].
For each of the company's 3-5 stated values, suggest which story from my bank best demonstrates that value, and what subtle reframing of the story emphasizes that value without changing the facts.
For each suggestion, mark when the reframing crosses into spin. The line is: am I emphasizing a real angle that was always there, or am I retrofitting a value onto a story it did not really fit?
Why this works: The reframing question is the right one because the same story does carry different angles. The “when does this cross into spin?” output is the safeguard that prevents you from sounding rehearsed in the wrong way. Interviewers can tell the difference between someone speaking authentically about a value they actually live and someone pattern-matching to the company’s stated values; this prompt keeps you on the right side of that line.
4. Technical screen prep
Inputs: my level of experience with each topic [paste, be candid with yourself].
For each topic:
– The 3 most likely interview questions at my level
– The expected solution shape (not the code; the approach)
– The trap question (the variation that tests whether I understood the underlying concept vs memorized the solution)
– The follow-up that probes depth ("walk me through your thinking")
– What to do if I do not know: the candid sentence I should say, then the second question I would ask the interviewer to make progress.
For each topic, suggest one specific practice problem I should solve before the interview, with a difficulty rating.
Why this works: The “what to do if I do not know” output is the part that distinguishes a candidate who passes technical screens from one who locks up. The candid sentence (“I have not worked with this specific pattern before, but here is how I would approach it”) followed by a clarifying question is the demonstrated pattern of someone who can work in a real engineering team. AI-generated solutions you memorized fall apart under the trap-question follow-up.
5. Case round / work-sample prep
Available time: [N hours].
First, walk through the case as I would in the room:
– The 3 questions I should ask the interviewer before starting
– The framework I should use, and why it fits this case
– The 5-step structure for the response, with time allocation
– The mistakes most candidates make on this case (over-framework, no synthesis, no recommendation, etc)
– The closing recommendation format the interviewer is hoping for
Then, given my background [paste], suggest one specific angle from my experience that would make my response distinctive. Do not invent experience I do not have.
Why this works: Case rounds reward structure over cleverness. The time-allocated 5-step structure is the difference between candidates who feel rushed and candidates who finish with a recommendation. The “3 questions to ask the interviewer” output is the part most candidates skip; clarifying questions before launching into the analysis is the single best signal of senior-candidate behavior.
6. The “any questions for us?” closer
Help me build a list of 8 strong questions and rank them by interviewer:
– 2 for the hiring manager
– 2 for a future peer
– 2 for a future direct report (if applicable)
– 2 for the recruiter / HR / final-round leadership
Each question should:
– Reveal something I cannot find on the company website
– Demonstrate that I have done real research
– Be a question I actually want the answer to (no theater)
– Not signal red-flag-checking too obviously ("why did the last person leave?" is asking, but ask it once and to one person, not in every round)
Tell me which question to save for the highest-stakes round.
Why this works: The closer is the part of the interview that you can fully control. Candidates who ask the same questions in every round signal that they did not research the company; candidates who ask different questions of different interviewers signal that they understand how organizations work. The “save the best for the highest-stakes round” output is the strategy most candidates miss.
7. The surprise question
For each question:
– Why this interviewer would ask it (their role, the company moment, the role's pressure points)
– The thing the question is actually testing (not the surface ask)
– The framework for a 60-90 second answer
– The trap I might walk into trying to answer
– The one-sentence opener that buys me 5 seconds to think
Why this works: Surprise questions are usually predictable in retrospect. The “what is the question actually testing” output is the realization that most “weird” questions are testing structured thinking, comfort with ambiguity, or how you handle stretch beyond your prepared answers. The “one-sentence opener that buys 5 seconds” is the practical bridge that lets you keep the room without freezing.
What is the worst thing you can do with AI in interview prep?
- Memorize AI-generated answers verbatim. The polish that sounds great in your head sounds rehearsed in the room. Use AI for structure; deliver in your own words.
- Let AI invent stories. Every story in the bank must be one you actually lived. AI is excellent at framing real experience and disastrous at fabricating it; the follow-up question is where the lie unravels.
- Skip the pressure-test (prompt 2). The polished primary answer is forgotten; the follow-up answer is remembered. Practicing only the polished part is practicing the easy part.
- Forget that the interviewer is reading your resume in real time. Cross-reference the resume bullet you cite in a story with what your resume actually says. AI sometimes paraphrases the bullet in a way that creates a small inconsistency.
- Use AI live in the interview. If the interview is virtual, this is increasingly easy and increasingly detected. The cost of being caught is much higher than the cost of saying “I do not know but here is how I would approach it.”
What if you are interviewing every week?
If you are in active search and running multiple interview cycles per week, the seven prompts above are a candidate for the ladder. Save the story-bank prompt as a Claude skill so you only do that work once. Save the value-translation and pressure-test prompts as skills that take the bank + a new company URL and produce a custom prep pack. Bundle them as interview-loop that you invoke before each round. The setup is 60-90 minutes. The time saved per interview is 1-2 hours of less-effective generic prep.
📊 The Prompt-to-Workflow Ladder
Tier 1: the prompts (this post). Tier 2: the skill (story-bank skill is the foundation). Tier 3: the plugin (interview-loop bundle). Tier 4: the workflow (prep pack auto-generates the night before the interview). When to climb →
What are common questions about AI interview prep?
Can interviewers tell when I prepped with AI?
Often, yes, if you used AI for the wrong thing. Memorized-sounding STAR answers, the same triple-adjective phrases your cover letter had, an opening sentence that mirrors AI defaults all signal “this was rehearsed with help.” Interviewers do not punish AI-assisted prep; they punish answers that sound machine-generated.
Should I disclose I used AI to prep?
Unprompted, no. If asked directly, yes (with the same framing you would use for a friend: AI helped me organize my thinking and pressure-test my stories; the experiences are mine). The disclosure is rarely volunteered and rarely held against you when offered.
How many stories should be in the bank?
Six is the sweet spot. Two leadership stories, two failure-or-learning stories, one conflict story, one accomplishment story. Most behavioral questions can be routed through one of these six. Twelve is too many to memorize cleanly; three leaves you exposed when the question does not match.
What if I lock up in the room?
Two practiced openers that buy you time: “Give me a moment to pick the right story” or “I want to make sure I answer the question you actually asked, can I check my understanding first?” Both are professional and both create the 10-15 seconds your brain needs to retrieve the story. Practice them.
Where do these prompts come from?
They are the interview-prep section of the larger AI Prompt Library. The Library has over 500 prompts across 33+ categories, including the full career stack: cover letters, resumes, interview prep, salary negotiation, and onboarding the first 90 days.
Sources to read next?
- “Cracking the Behavioral Interview” methodology (industry-standard STAR-deep)
- Harvard Business Review on interview research
- The Muse interview advice library
- SHRM hiring-process research
- Job interview (Grokipedia)
✏️ Before the interview
Polished AI-generated answers are flagged as rehearsed within 60 seconds. Our editing pass for written work applies to spoken answers too: How to Edit AI Out of Your Writing → The full 29-pattern catalog from Wikipedia's “Signs of AI writing” guide is documented there for any written prep (cover letter, thank-you note, follow-up email) that supports the interview.
The AI Prompt Library · $39
over 500 tested prompts including the full career stack
The seven interview prompts above are a free preview. The full Library has the full career stack: cover letters, resumes, interview prep, salary negotiation, onboarding the first 90 days, and the harder situations like career pivots, returning from a gap, and after-layoff outreach.
1-on-1 Custom AI Tutorial with James · $99
Build your story bank live, then pressure-test it
A 1-hour private call. We build the story bank live, run the pressure-test prompt on your three weakest stories, and build the interview-loop skill so you walk into next week's interviews with reusable prep that takes 15 minutes instead of 90.
Get Smarter About AI Every Morning
Free daily newsletter. Built for people who want to use AI well, not chase every model.
Free forever. Unsubscribe anytime.
You may also like
- Best AI Prompts for Cover Letters — the front of the job-search funnel
- Best AI Prompts for Resume Writing — what gets you in the door
- Best Prompts for Salary Negotiation — the conversation that pays for itself
- How to Edit AI Out of Your Writing — strip AI patterns before you publish
- Prompt to Workflow: The AI Ladder — when to climb from prompt to workflow
Two ways to go further
The AI Prompt Library
1,000+ ready-to-use prompts for Claude, ChatGPT, and Gemini. Stop staring at a blank box.
Get it for $39 →2-Hour Live AI Crash Course
A private, beginner-friendly session across Claude, ChatGPT, Gemini, and the wider landscape.
Book for $125 →