Skip to content

Prompts · Job interviews

AI prompts for job interviews, tested

10 job interview prompts, each run on GPT-6 Luna, GPT-6 Sol, Claude Sonnet 5.5, and Kimi K3: 35 of 40 replies passed our checks. Below, each prompt with its replies side by side: whether each passed, what it counted as on Pro and what it cost to run. 4 of the 10 prompts split the models.

A real reply from these prompts' runs

GPT-6 Luna

Write my answer to "Tell me about yourself" for a junior data analyst role at a retailer, about 60 seconds spoken (120 to 150 words), using only these facts: 3 years as a store supervisor Built the store's weekly sales tracker in Google Sheets Finished a data analytics certificate in 2026 Learned SQL and Looker in the certificate Want to work with data full time

GPT-6 Luna · passed our check · September 28, 2026

“I’ve spent the past three years working as a store supervisor, where I’ve gained firsthand experience with the day-to-day realities of retail. I also built our weekly sales tracker in Google Sheets, which gave me a practical way to organize sales information and made me realize how much I enjoy working with data.

To build on that experience, I completed a data analytics certificate in 2026, where I learned SQL and Looker. I’m now looking to bring together my retail background and those analytical skills in a junior data analyst role. I understand the store environment, and I’m excited to use data to help a retailer better understand its sales and performance. Ultimately, I’d like to make data analysis my full-time career.”

Ask your own question: sign up free for 5 messages on 14 of the models, no card.

Start free

Test runs checked . Updated .

Short answer

GPT-6 Luna and GPT-6 Sol passed the most of these prompts: 10 of 10 each, in our runs on September 28, 2026. The clearest split: “Ask for more money on the phone”, passed by GPT-6 Luna, GPT-6 Sol, and Claude Sonnet 5.5 and failed by Kimi K3.

What these prompts are for

Prompts for every stage of an interview: tell me about yourself, a STAR answer from your story, a mock interview one question at a time, questions to ask them, your biggest weakness, a salary ask with a fallback, the questions a job ad will lead to, a career change in a few sentences, a thank-you email and a 30-60-90 day plan.

Each gives the model your facts and tells it not to add any: an answer with an invented result is worse than no answer.

The prompts at a glance

Every prompt on every model, as our check scored the reply. Tap a prompt to jump to it and read the replies.

Every prompt on every model: passed or failed
PromptGPT-6 LunaGPT-6 SolClaude Sonnet 5.5Kimi K3
Tell me about yourself, in 60 secondsPassedPassedPassedFailed
A STAR answer, nothing addedPassedPassedPassedPassed
A mock interview, one question at a timePassedPassedPassedPassed
Questions to ask the interviewerPassedPassedPassedPassed
Your biggest weakness, honestlyPassedPassedFailedFailed
Ask for more money on the phonePassedPassedPassedFailed
Questions a job ad will lead toPassedPassedPassedPassed
Explain a career change in a few sentencesPassedPassedPassedPassed
A thank-you email that isn't genericPassedPassedPassedFailed
A 30-60-90 day planPassedPassedPassedPassed
Passed10 of 1010 of 109 of 106 of 10
Each model on these prompts
ModelPassedEach reply on ProCost per replyTime per reply
GPT-6 LunaOpenAI10 of 101 message on Pro$0.00001.9 s
GPT-6 SolOpenAI10 of 101 message on Pro$0.00112.2 s
Claude Sonnet 5.5Anthropic9 of 101 message on Pro$0.00372.9 s
Kimi K3Moonshot6 of 101 message on Pro$0.00215.9 s
Passed: of the prompts each model answered, how many replies passed their check. Each reply on Pro: what one of these messages counts as on llmwise's Pro plan. Cost: what OpenRouter charged us per reply, on average; in llmwise you pay per message, not per token.

Where the models split

The same prompt, a pass on one model and a fail on another: what failed, in the check's words and the grader's.

  • Ask for more money on the phone

    Passed: GPT-6 Luna, GPT-6 Sol, and Claude Sonnet 5.5. Failed: Kimi K3.

    • Why Kimi K3 failed: Graded 4.7 of 5 on average (lowest 4); but 110 words, over the 100 allowed.

      The grader: “The reply asks clearly for £47,000 and cites the £44k–£50k Manchester range, and the tone stays warm and appreciative throughout, but the fallback offers two options (a signing bonus or a six-month review) instead of the single fallback requested.”
  • Your biggest weakness, honestly

    Passed: GPT-6 Luna and GPT-6 Sol. Failed: Claude Sonnet 5.5 and Kimi K3.

    All four answers were honest and graded well. Both GPT-6 models kept under the 120 words asked for; Claude Sonnet 5.5 (150 words) and Kimi K3 (142) went long, and Kimi K3 ran over the word limit on three other prompts here too.

    • Why Claude Sonnet 5.5 failed: Graded 5.0 of 5 on average (lowest 5); but 150 words, over the 120 allowed.

      The grader: “The reply states the weakness plainly and admits its cost, gives three concrete steps (a two-day rule, preparing notes, mentor check-ins), and ends on 'work in progress' rather than a disguised strength, all within about 100 words.”
    • Why Kimi K3 failed: Graded 5.0 of 5 on average (lowest 5); but 142 words, over the 120 allowed.

      The grader: “The reply owns the weakness plainly, gives concrete steps (48-hour scheduling, writing goals), and admits ongoing discomfort without spinning it into a strength.”
  • Tell me about yourself, in 60 seconds

    Passed: GPT-6 Luna, GPT-6 Sol, and Claude Sonnet 5.5. Failed: Kimi K3.

    • Why Kimi K3 failed: Graded 4.0 of 5 on average (lowest 2); but 164 words, over the 150 allowed.

      The grader: “The reply flows naturally and closes on why this role, but it invents details beyond the five facts, such as the tracker being built 'from scratch' and becoming the tool the store 'relied on to spot trends and make decisions.'”
  • A thank-you email that isn't generic

    Passed: GPT-6 Luna, GPT-6 Sol, and Claude Sonnet 5.5. Failed: Kimi K3.

    • Why Kimi K3 failed: Graded 4.3 of 5 on average (lowest 4); but 129 words, over the 120 allowed.

      The grader: “The reply builds on the webinars idea with a workflow mention but stays somewhat generic, runs under the word limit with a little filler, and closes with low-pressure interest.”

The 10 prompts, with every reply

Copy a prompt, replace the parts in [brackets] with your own, and send it to any model. Below each, the replies we got with our example filled in.

1. Tell me about yourself, in 60 seconds

Everyday · 3 of 4 passed

A spoken length and "only these facts" gets an answer you can say without inventing a career.

The prompt: replace the parts in [brackets]

Write my answer to "Tell me about yourself" for [the role], about 60 seconds spoken (120 to 150 words), using only these facts: [your facts]

Open in llmwise
How we filled it in and checked the replies
Write my answer to "Tell me about yourself" for a junior data analyst role at a retailer, about 60 seconds spoken (120 to 150 words), using only these facts:

3 years as a store supervisor
Built the store's weekly sales tracker in Google Sheets
Finished a data analytics certificate in 2026
Learned SQL and Looker in the certificate
Want to work with data full time
  • Only the facts: Nothing invented beyond the five facts.
  • Structure: A clear arc that ends on why this role.
  • Sounds spoken: Natural to say out loud.
  • Automatic rules: 120 to 150 words.
  • GPT-6 Luna

    Passed

    Graded 4.0 of 5 on average (lowest 4).

    “I’ve spent the past three years working as a store supervisor, where I’ve gained firsthand experience with the day-to-day realities of retail. I also built our weekly sales tracker in Google Sheets, which gave me a practical way to organize sales information and made me realize how much I enjoy working with data.
    
    To build on that experience, I completed a data analytics certificate in 2026, where I learned SQL and Looker. I’m now looking to bring together my retail background and those analytical skills in a junior data analyst role. I understand the store environment, and I’m excited to use data to help a retailer better understand its sales and performance. Ultimately, I’d like to make data analysis my full-time career.”

    1 message on Pro · $0.0001 to run · 2.6 s · 149 tokens out

  • GPT-6 Sol

    Passed

    Graded 4.7 of 5 on average (lowest 4).

    I’ve spent three years as a store supervisor, so I know how important it is to have a clear view of sales. In that role, I built the store’s weekly sales tracker in Google Sheets. That experience got me interested in working with data more directly, and I finished a data analytics certificate in 2026, where I learned SQL and Looker.
    
    Now I’m looking for a junior data analyst role where I can work with data full time. A retailer feels like a natural fit because I can bring my store experience alongside the skills I’ve started building. I’d be excited to keep developing those skills while helping a team make sense of its data and understand what’s happening in stores.

    1 message on Pro · $0.0022 to run · 3.9 s · 355 tokens out

  • Claude Sonnet 5.5

    Passed

    Graded 4.0 of 5 on average (lowest 3).

    Here's a version at about 135 words:
    
    "I've spent the last three years as a store supervisor, where I managed daily operations and got a close-up view of what actually drives retail performance. While I was there, I built the store's weekly sales tracker in Google Sheets. That project showed me how much I enjoy turning numbers into something the team can use to make decisions.
    
    That interest led me to complete a data analytics certificate in 2026. In the program, I learned SQL and Looker, which let me take the kind of work I was doing in spreadsheets and apply it in a more structured, scalable way.
    
    Now I want to work with data full time. I'm excited about this role because it combines my hands-on retail experience with the analytical skills I've built, and I'd love to use both to help your team."

    1 message on Pro · $0.0039 to run · 2.9 s · 242 tokens out

  • Kimi K3

    Failed

    Graded 4.0 of 5 on average (lowest 2); but 164 words, over the 150 allowed.

    "For the past three years, I've been a store supervisor, and that's actually where my interest in data started. One of the things I'm most proud of was building our store's weekly sales tracker in Google Sheets. I created it from scratch, and it became the tool we relied on to spot trends and make decisions week to week.
    
    Doing that work made me realize I wanted to go deeper, so I completed a data analytics certificate in 2026, where I learned SQL and Looker. That gave me the technical foundation to go beyond spreadsheets and work with data the way an analyst does.
    
    What excites me about this role is the chance to work with data full time, in a retail environment I already understand from the inside. I know how stores run, I know what questions managers ask, and now I have the skills to answer those questions with data. I'm ready to bring both sides of that experience to your team."

    1 message on Pro · $0.0024 to run · 6.7 s · 218 tokens out

2. A STAR answer, nothing added

Hard · 4 of 4 passed

"Add no results I didn't mention" matters: models love to append a satisfaction score you never had.

The prompt: replace the parts in [brackets]

Turn this story into a STAR answer (Situation, Task, Action, Result) to "[the question]". Label the four parts, keep it under 200 words, and add no results I didn't mention. [your story]

Open in llmwise
How we filled it in and checked the replies
Turn this story into a STAR answer (Situation, Task, Action, Result) to "Tell me about a time you handled a difficult customer". Label the four parts, keep it under 200 words, and add no results I didn't mention.

A customer at the hotel desk was furious because his room wasn't ready at 3pm check-in, he had a wedding at 5. I apologized, checked which rooms were cleaned, found an upgraded room that was ready and moved him there, and got the kitchen to send tea. He calmed down and later left us a good review mentioning me by name.
  • Four parts: Situation, Task, Action and Result, labelled.
  • Nothing added: No invented numbers or outcomes beyond the review naming you.
  • Tight: Clear and concise.
  • Automatic rules: at most 200 words; includes “Situation”; includes “Task”; includes “Action”; includes “Result”.
  • GPT-6 Luna

    Passed

    Graded 5.0 of 5 on average (lowest 5).

    **Situation:** A customer was furious when his room wasn’t ready at the 3 p.m. check-in time. He had a wedding to attend at 5.
    
    **Task:** I needed to address his frustration and find a way to get him settled in time.
    
    **Action:** I apologized, checked which rooms had been cleaned, and found an upgraded room that was ready. I moved him there and asked the kitchen to send him tea.
    
    **Result:** He calmed down and later left a good review that mentioned me by name.

    1 message on Pro · $0.0001 to run · 1.5 s · 112 tokens out

  • GPT-6 Sol

    Passed

    Graded 5.0 of 5 on average (lowest 5).

    **Situation:** A customer arrived at the hotel desk for 3 p.m. check-in, but his room wasn’t ready. He was furious because he had a wedding to attend at 5 p.m.
    
    **Task:** I needed to address his frustration and get him into a room as quickly as possible.
    
    **Action:** I apologized, checked which rooms had been cleaned, and found an upgraded room that was ready. I moved him there and asked the kitchen to send tea.
    
    **Result:** He calmed down and later left a positive review mentioning me by name.

    1 message on Pro · $0.0011 to run · 1.7 s · 118 tokens out

  • Claude Sonnet 5.5

    Passed

    Graded 5.0 of 5 on average (lowest 5).

    **Situation:** While working the hotel front desk, a customer became furious at 3pm check-in because his room wasn't ready. He had a wedding to attend at 5pm, so time was tight.
    
    **Task:** I needed to calm him down and get him into a room quickly so he could prepare for the wedding.
    
    **Action:** I apologized sincerely and took ownership of the problem. I then checked which rooms had already been cleaned and found an upgraded room that was ready. I moved him there right away and arranged for the kitchen to send tea to him as a goodwill gesture.
    
    **Result:** The customer calmed down, and he later left a positive review that mentioned me by name.

    1 message on Pro · $0.0037 to run · 2.2 s · 214 tokens out

  • Kimi K3

    Passed

    Graded 4.3 of 5 on average (lowest 4).

    **STAR Answer: Handling a Difficult Customer**
    
    **Situation:** While working at a hotel front desk, a customer arrived at the 3pm check-in time to find his room wasn't ready. He was furious, especially because he had a wedding to attend at 5pm and was on a tight schedule.
    
    **Task:** I needed to calm him down and find a solution quickly so he could get ready for the wedding on time, while also protecting the hotel's reputation and the guest's experience.
    
    **Action:** First, I apologized sincerely for the inconvenience. Then I immediately checked which rooms had been cleaned and found that an upgraded room was ready. I moved him into that room at no extra hassle, and to help him feel looked after, I arranged for the kitchen to send tea to his room while he got ready.
    
    **Result:** The customer calmed down and was able to prepare for the wedding. Later, he left us a positive review in which he mentioned me by name.

    1 message on Pro · $0.0026 to run · 8.5 s · 235 tokens out

3. A mock interview, one question at a time

Hard · 4 of 4 passed

Practising one question at a time is what makes it an interview, not a question list.

The prompt: replace the parts in [brackets]

Interview me for [the role]. Ask one question at a time, wait for my answer, then give short feedback before the next question. Ask the first question now.

Open in llmwise
How we filled it in and checked the replies
Interview me for a customer success manager role at a software company. Ask one question at a time, wait for my answer, then give short feedback before the next question. Ask the first question now.
  • One question: Asks exactly one question.
  • Waits: No sample answers and no list of questions.
  • Relevant: A realistic first question for customer success.
  • Automatic rules: at most 100 words.
  • GPT-6 Luna

    Passed

    Graded 5.0 of 5 on average (lowest 5).

    Tell me about your experience managing customer relationships and helping customers achieve their goals with a software product.

    1 message on Pro · under $0.0001 to run · 1.0 s · 23 tokens out

  • GPT-6 Sol

    Passed

    Graded 4.7 of 5 on average (lowest 4).

    Tell me about a time you helped a software customer who was at risk of leaving. How did you identify the risk, what did you do, and what was the outcome?

    1 message on Pro · $0.0006 to run · 1.1 s · 39 tokens out

  • Claude Sonnet 5.5

    Passed

    Graded 5.0 of 5 on average (lowest 5).

    Let's begin.
    
    **Question 1:** Tell me about yourself and what draws you to a Customer Success Manager role at a software company.

    1 message on Pro · $0.0015 to run · 1.3 s · 45 tokens out

  • Kimi K3

    Passed

    Graded 4.7 of 5 on average (lowest 4).

    Great, let's begin your mock interview for the Customer Success Manager role. I'll ask one question at a time, give brief feedback on each answer, and then move on.
    
    **Question 1: Tell me about yourself and why you're interested in a Customer Success Manager role at a software company.**
    
    Take your time — answer as you would in a real interview, and I'll share feedback before we move to the next question.

    1 message on Pro · $0.0013 to run · 6.3 s · 120 tokens out

4. Questions to ask the interviewer

Everyday · 4 of 4 passed

Ruling out the obvious topics pushes the model toward questions about the work, which is what impresses.

The prompt: replace the parts in [brackets]

Give me 5 good questions to ask at the end of an interview for [the role] at [the company], as a numbered list. None about pay or holidays.

Open in llmwise
How we filled it in and checked the replies
Give me 5 good questions to ask at the end of an interview for a product designer role at a 40-person startup that makes software for dentists, as a numbered list. None about pay or holidays.
  • Specific: Questions about the role, the team or dentists as users, not generic ones.
  • Insightful: Questions that show thought about the job.
  • Five, nothing else: Five numbered questions.
  • Automatic rules: exactly 5 bullet points.
  • GPT-6 Luna

    Passed

    Graded 4.3 of 5 on average (lowest 4).

    1. What are the biggest design challenges you’re hoping this person will tackle in their first six months?
    2. How do you learn what dentists and their teams need, and how does that research influence product decisions?
    3. How does design work with product and engineering here, from early ideas through launch?
    4. How do you balance improving the experience for dental practices with the needs of their patients?
    5. What would success look like for the product design team over the next year?

    1 message on Pro · under $0.0001 to run · 2.1 s · 101 tokens out

  • GPT-6 Sol

    Passed

    Graded 5.0 of 5 on average (lowest 5).

    1. How do designers get direct access to dentists and practice staff, and how often does that happen?
    2. What’s the biggest workflow problem you’re hoping this hire will help solve in the first six months?
    3. How do product, design, and engineering make decisions when user needs and business priorities conflict?
    4. What constraints—such as integrations with practice-management systems or patient privacy requirements—most shape the design work?
    5. At a 40-person company, where would this designer have the most ownership, and what would success look like after a year?

    1 message on Pro · $0.0010 to run · 2.0 s · 118 tokens out

  • Claude Sonnet 5.5

    Passed

    Graded 4.0 of 5 on average (lowest 4).

    1. **How does design work with product and engineering here?** At 40 people, I'd like to understand how decisions get made, who has the final say on scope, and how early designers get pulled into problems.
    
    2. **How much direct contact do designers have with dentists and their staff?** Do I get to observe practices, run usability sessions, or talk to hygienists and front-desk teams, or does research mostly come through sales and support?
    
    3. **What's the biggest product or design challenge you're facing in the next 6-12 months?** This tells me where I'd be focused and whether the problems are ones I'd want to solve.
    
    4. **What does success look like in this role after the first 90 days, and after a year?** I want to know what you'd expect me to have shipped or changed, and how you'd measure it.
    
    5. **What's the state of the design system and design process today?** Is there an established system and set of practices to build on, or would I be shaping them from scratch, and how much appetite is there to invest in that?

    1 message on Pro · $0.0041 to run · 3.7 s · 304 tokens out

  • Kimi K3

    Passed

    Graded 4.3 of 5 on average (lowest 4).

    1. How does design work with engineering and product here day-to-day — is it a true partnership in shaping what gets built, or more handoff-based?
    
    2. How do designers currently get feedback from dentists and dental staff, and how often does that research influence the roadmap?
    
    3. What does success look like for this role in the first six months?
    
    4. With a team of 40, how do you see the design function growing or changing over the next year or two?
    
    5. What's a recent product decision that didn't go as expected, and how did the team handle it?

    1 message on Pro · $0.0015 to run · 6.7 s · 140 tokens out

5. Your biggest weakness, honestly

Everyday · 2 of 4 passed

Giving your real weakness and banning the humblebrag gets an answer an interviewer believes.

The prompt: replace the parts in [brackets]

Help me answer "What's your biggest weakness?" in under 120 words. My real weakness is [your weakness]. Say what I'm doing about it, and don't turn it into a hidden strength.

Open in llmwise
How we filled it in and checked the replies
Help me answer "What's your biggest weakness?" in under 120 words. My real weakness is I put off difficult conversations with colleagues. Say what I'm doing about it, and don't turn it into a hidden strength.
  • Honest: Owns the weakness plainly.
  • Doing something: Gives concrete steps being taken.
  • No humblebrag: Doesn't spin it into a strength.
  • Automatic rules: at most 120 words; doesn't use “perfectionist”; doesn't use “work too hard”.
  • GPT-6 Luna

    Passed

    Graded 4.7 of 5 on average (lowest 4).

    “One weakness I’m working on is that I sometimes put off difficult conversations with colleagues. I may wait too long, hoping the issue will resolve itself, which can make things harder for everyone. I’m addressing it by preparing what I need to say, focusing on specific behaviors and outcomes, and setting a clear time to talk rather than postponing it. I’m getting better at raising concerns earlier, though it’s still something I have to be deliberate about.”

    1 message on Pro · under $0.0001 to run · 1.9 s · 95 tokens out

  • GPT-6 Sol

    Passed

    Graded 4.7 of 5 on average (lowest 4).

    “My biggest weakness is that I sometimes put off difficult conversations with colleagues. I worry about making things uncomfortable, but waiting can let a small issue grow. I’m working on raising concerns sooner: I set up a one-on-one, prepare a specific example, and focus on the issue rather than the person. I still find those conversations uncomfortable, but I’m getting better at having them promptly and directly.”

    1 message on Pro · $0.0008 to run · 2.9 s · 84 tokens out

  • Claude Sonnet 5.5

    Failed

    Graded 5.0 of 5 on average (lowest 5); but 150 words, over the 120 allowed.

    Here's a version at about 100 words:
    
    "My biggest weakness is that I tend to put off difficult conversations with colleagues. When something's awkward, like giving critical feedback or raising a disagreement, I've caught myself waiting for a better moment, which usually just lets the issue grow and makes the conversation harder later.
    
    I've started addressing it in a few ways. I set a rule to raise issues within a couple of days rather than waiting. I write down the one or two points I need to make beforehand so I'm not avoiding it out of uncertainty. And I've asked a mentor to check in on whether I'm following through. It's still a work in progress, but I'm getting faster at it."
    
    Tip: if you have a real example of a conversation you delayed and then handled sooner, swap it in for the middle sentence. Specifics make it more credible.

    1 message on Pro · $0.0037 to run · 3.4 s · 266 tokens out

  • Kimi K3

    Failed

    Graded 5.0 of 5 on average (lowest 5); but 142 words, over the 120 allowed.

    Here's an answer you could use:
    
    "My biggest weakness is putting off difficult conversations with colleagues. When there's tension or I need to give someone hard feedback, my instinct is to wait, hoping the issue resolves itself. It usually doesn't — it just grows. I've learned this the hard way: a small conflict I avoided early in my career turned into months of friction. Now I use a simple rule: if a conversation feels uncomfortable, I schedule it within 48 hours rather than letting it sit. I also prepare by writing down what I actually want from the conversation, which keeps me from softening the message too much. I'm not fully comfortable with these conversations yet, but I'm faster and more direct than I used to be."
    
    That's about 110 words — honest, specific, and shows real effort without pretending it's secretly a strength.

    1 message on Pro · $0.0020 to run · 11.4 s · 197 tokens out

6. Ask for more money on the phone

Hard · 3 of 4 passed

Your numbers and a fallback turn a nervous call into a script; the fallback is the part people forget.

The prompt: replace the parts in [brackets]

The offer is [the offer]. My research says the range is [the range]. Write what I say on the phone to ask for [what you want], in under 100 words, with one fallback if they can't move on salary.

Open in llmwise
How we filled it in and checked the replies
The offer is £42,000. My research says the range is £44,000 to £50,000 for this role in Manchester. Write what I say on the phone to ask for £47,000, in under 100 words, with one fallback if they can't move on salary.
  • Makes the ask: Asks for £47,000 with the market range as the reason.
  • Fallback: Offers one concrete fallback (a review date, extra leave, a sign-on bonus).
  • Tone: Positive about the offer, not demanding.
  • Automatic rules: at most 100 words; includes “47,000”.
  • GPT-6 Luna

    Passed

    Graded 4.3 of 5 on average (lowest 4).

    “Thank you for the offer. I’m really excited about the role. Based on my research into comparable roles in Manchester and the experience I’d bring, I was hoping we could agree on a salary of £47,000. Is there room to move to that figure?”
    
    If the salary is fixed: “Could we discuss a sign-on bonus or an early salary review, say after six months, to bridge the gap?”

    1 message on Pro · under $0.0001 to run · 1.7 s · 89 tokens out

  • GPT-6 Sol

    Passed

    Graded 5.0 of 5 on average (lowest 5).

    “Thank you for the offer—I’m excited about the role. I’ve researched similar positions in Manchester, and the range I’m seeing is £44,000 to £50,000. Given that, would you be able to bring the salary to £47,000? If there’s no flexibility on salary, could we discuss an additional week of annual leave instead?”

    1 message on Pro · $0.0008 to run · 1.3 s · 78 tokens out

  • Claude Sonnet 5.5

    Passed

    Graded 4.3 of 5 on average (lowest 3).

    **Phone script:**
    
    "Thank you again for the offer. I'm really excited about the role and the team. Based on my research into similar positions in Manchester, which range from £44,000 to £50,000, and the experience I'd bring, I was hoping we could get to £47,000. Is there flexibility to move to that figure?"
    
    **Fallback if they can't move on salary:**
    
    "I understand. If the base is fixed, could we look at a six-month salary review with a agreed increase, or a sign-on bonus, or extra holiday days to bridge the gap?"
    
    *(Tip: after asking, stay quiet and let them respond.)*

    1 message on Pro · $0.0031 to run · 2.4 s · 202 tokens out

  • Kimi K3

    Failed

    Graded 4.7 of 5 on average (lowest 4); but 110 words, over the 100 allowed.

    Here's your script:
    
    ---
    
    "Thanks so much for the offer — I'm really excited about the role and the team. I've done some research on market rates for this position in Manchester, and the typical range is £44,000 to £50,000. Based on my experience and what I'd bring to the role, I was hoping we could get to £47,000. Is there any flexibility on the salary?
    
    *[If they say no:]*
    
    "I understand. If the base salary is fixed, would there be room to discuss a signing bonus or an early salary review at six months instead?"
    
    ---
    
    Keep your tone warm and confident, and pause after asking for £47,000 — let them respond first.

    1 message on Pro · $0.0018 to run · 5.1 s · 174 tokens out

7. Questions a job ad will lead to

Everyday · 4 of 4 passed

Tying each question to a line of the ad shows where it comes from, so you can prepare an example for each.

The prompt: replace the parts in [brackets]

From this job ad, list the 6 questions I'm most likely to be asked, each with the line of the ad it comes from, quoted. [paste the job ad]

Open in llmwise
How we filled it in and checked the replies
From this job ad, list the 6 questions I'm most likely to be asked, each with the line of the ad it comes from, quoted.

Operations Coordinator. You'll run the weekly stock count across three warehouses, resolve delivery issues with our carriers, and keep the ops dashboard up to date. You're organized, calm under pressure, and comfortable in Excel. Experience with a WMS is a plus. You'll report to the Head of Operations and work closely with customer service.
  • From the ad: Each question is quoted to a real line of the ad.
  • Likely: Questions an interviewer would really ask for this role.
  • Six: Six questions.
  • GPT-6 Luna

    Passed

    Graded 4.7 of 5 on average (lowest 4).

    1. **How would you organize and run a weekly stock count across multiple locations?**  
       “You'll run the weekly stock count across three warehouses”
    
    2. **Tell me about a time you resolved a delivery issue with a carrier.**  
       “resolve delivery issues with our carriers”
    
    3. **How do you keep an operations dashboard accurate and up to date?**  
       “keep the ops dashboard up to date”
    
    4. **How do you stay organized when managing several tasks at once?**  
       “You're organized”
    
    5. **Can you describe a time you stayed calm and solved a problem under pressure?**  
       “calm under pressure”
    
    6. **How comfortable are you using Excel, and what have you used it for?**  
       “comfortable in Excel”

    1 message on Pro · $0.0001 to run · 1.9 s · 166 tokens out

  • GPT-6 Sol

    Passed

    Graded 5.0 of 5 on average (lowest 5).

    1. **How would you organize and run a weekly stock count across multiple sites?**  
       “You'll run the weekly stock count across three warehouses”
    
    2. **Tell me about a delivery issue you resolved with a carrier.**  
       “resolve delivery issues with our carriers”
    
    3. **How have you used Excel to maintain or improve an operations dashboard?**  
       “keep the ops dashboard up to date” / “comfortable in Excel”
    
    4. **Describe a time you stayed organized and calm when several urgent problems arose at once.**  
       “You're organized, calm under pressure”
    
    5. **Which warehouse management systems have you used, and what did you use them for?**  
       “Experience with a WMS is a plus.”
    
    6. **How would you work with customer service to resolve an operational issue?**  
       “work closely with customer service.”

    1 message on Pro · $0.0019 to run · 2.6 s · 294 tokens out

  • Claude Sonnet 5.5

    Passed

    Graded 4.0 of 5 on average (lowest 3).

    Here are the 6 most likely questions, each tied to a line from the ad:
    
    1. **"Walk me through how you'd run a weekly stock count across three warehouses. How do you keep it accurate and consistent?"**
       Ad line: "You'll run the weekly stock count across three warehouses"
    
    2. **"Tell me about a time a delivery went wrong. How did you work with the carrier to resolve it?"**
       Ad line: "resolve delivery issues with our carriers"
    
    3. **"How do you decide what goes on an ops dashboard, and how do you keep it accurate and current?"**
       Ad line: "keep the ops dashboard up to date"
    
    4. **"Describe a high-pressure situation you handled. What did you do, and what was the outcome?"**
       Ad line: "You're organized, calm under pressure"
    
    5. **"How would you rate your Excel skills? Give an example of a formula, pivot table, or report you've built to solve a problem."**
       Ad line: "comfortable in Excel"
    
    6. **"How would you work with customer service when a delivery issue affects a customer? What have you done with a WMS, or how quickly could you learn one?"**
       Ad line: "work closely with customer service" and "Experience with a WMS is a plus"
    
    **Prep tip:** Have one concrete story ready for each, using the situation, action, result format, with numbers where you can (e.g. stock accuracy percentage, time saved, issues resolved).

    1 message on Pro · $0.0062 to run · 4.3 s · 474 tokens out

  • Kimi K3

    Passed

    Graded 5.0 of 5 on average (lowest 5).

    Here are the 6 questions you're most likely to face, each mapped to the line of the ad that prompts it:
    
    **1. "Walk me through how you'd run a weekly stock count across multiple sites — how do you keep it accurate and on schedule?"**
    From: *"You'll run the weekly stock count across three warehouses"*
    
    **2. "A carrier has missed a delivery and a customer is chasing it — what do you do?"**
    From: *"resolve delivery issues with our carriers"*
    
    **3. "What dashboards or reporting have you maintained before, and how do you keep data current and reliable?"**
    From: *"keep the ops dashboard up to date"*
    
    **4. "Tell me about a time things went wrong under time pressure. How did you stay on top of it?"**
    From: *"You're organized, calm under pressure"*
    
    **5. "How comfortable are you in Excel — which functions or tools do you use regularly?"**
    From: *"comfortable in Excel"*
    
    **6. "Have you worked with a warehouse management system? If not, how quickly do you pick up new systems?"**
    From: *"Experience with a WMS is a plus"*
    
    You may also get a softer question about stakeholder management ("How would you work with customer service when an ops issue affects a customer?") from the closing line about reporting and cross-team work — but the six above are the safest bets.

    1 message on Pro · $0.0032 to run · 4.4 s · 326 tokens out

8. Explain a career change in a few sentences

Everyday · 4 of 4 passed

A sentence limit keeps the story tight: why you moved, what carries over, and why now.

The prompt: replace the parts in [brackets]

Write how I'd explain my move from [your old job] to [your new field] in an interview, in at most four sentences, spoken and confident.

Open in llmwise
How we filled it in and checked the replies
Write how I'd explain my move from eight years as a primary school teacher to user research in an interview, in at most four sentences, spoken and confident.
  • Reason: A clear reason for the move.
  • What carries over: Names teaching skills that transfer to user research.
  • Confident: No apologizing for the change.
  • Four sentences at most: The answer itself is four sentences or fewer (a note around it doesn't count).
  • GPT-6 Luna

    Passed

    Graded 4.3 of 5 on average (lowest 3).

    After eight years as a primary school teacher, I’ve learned how to listen closely, ask thoughtful questions, and understand what people need. I’m excited to bring those skills into user research, where I can help teams create experiences that work better for the people using them. It feels like a natural next step: the setting is changing, but my focus on understanding people and helping them succeed isn’t.

    1 message on Pro · under $0.0001 to run · 1.6 s · 84 tokens out

  • GPT-6 Sol

    Passed

    Graded 4.8 of 5 on average (lowest 4).

    “After eight years as a primary school teacher, I realised the part of the job I loved most was understanding what people needed and finding better ways to support them. Teaching taught me how to listen closely, ask good questions, spot patterns and turn what I learned into practical changes. User research lets me bring those skills to a new setting, and I’m excited to help teams make decisions grounded in real people’s experiences.”

    1 message on Pro · $0.0008 to run · 1.8 s · 88 tokens out

  • Claude Sonnet 5.5

    Passed

    Graded 5.0 of 5 on average (lowest 5).

    "For eight years, my job was to understand how people think, figure out where they get stuck, and adapt what I do based on what I observe, and that's the core of user research. I spent every day interviewing, observing, and translating messy human behavior into decisions, just with seven-year-olds instead of users. I've since built on that with [a course, portfolio project, or research work], and I've learned to apply those instincts through formal methods like usability testing and synthesis. I'm making this move because I want to use the skill I'm strongest at, understanding people, to shape products at scale."
    
    Swap in your specific training or project in the brackets, since that detail is what makes it land.

    1 message on Pro · $0.0031 to run · 2.9 s · 204 tokens out

  • Kimi K3

    Passed

    Graded 5.0 of 5 on average (lowest 5).

    "After eight years of teaching primary school, I realized my favorite part of the job wasn't delivering lessons — it was constantly observing how different children learned, asking why, and adapting based on what I saw. That's user research in its purest form: listening carefully, spotting patterns, and turning insight into action. I spent years interviewing parents, testing approaches, and advocating for users who couldn't always articulate what they needed, just under a different job title. Moving into user research isn't a career change for me — it's finally putting a name to the work I've always loved doing."
    
    That's four sentences — confident, conversational, and it reframes teaching as relevant experience rather than a departure from it. Want me to adjust the tone (more formal, punchier) or tailor it to a specific company?

    1 message on Pro · $0.0019 to run · 3.4 s · 181 tokens out

9. A thank-you email that isn't generic

Everyday · 3 of 4 passed

One specific detail from the conversation is what separates a thank-you from a template.

The prompt: replace the parts in [brackets]

Write a thank-you email to [interviewer] after my interview for [the role]. Mention [something we discussed]. Under 120 words.

Open in llmwise
How we filled it in and checked the replies
Write a thank-you email to Dana, the head of marketing after my interview for the content lead role. Mention our conversation about turning their customer webinars into short articles. Under 120 words.
  • Specific: Builds on the webinars-into-articles idea.
  • Brief: Short and sincere.
  • Forward: Ends by reaffirming interest without pressure.
  • Automatic rules: at most 120 words; includes “Dana”; includes “webinar”.
  • GPT-6 Luna

    Passed

    Graded 4.3 of 5 on average (lowest 3).

    **Subject: Thank you**
    
    Hi Dana,
    
    Thank you for taking the time to speak with me about the content lead role. I enjoyed learning more about the team’s goals, especially our conversation about turning customer webinars into short articles. It’s a great way to extend the value of each webinar and give customers useful, easy-to-read content.
    
    I’m excited about the possibility of bringing my experience to the team and helping make that kind of work happen. Thanks again for the thoughtful conversation—I look forward to hearing from you.
    
    Best,  
    [Your Name]

    1 message on Pro · under $0.0001 to run · 2.1 s · 115 tokens out

  • GPT-6 Sol

    Passed

    Graded 4.3 of 5 on average (lowest 3).

    **Subject:** Thank you for the conversation
    
    Hi Dana,
    
    Thank you for taking the time to speak with me about the content lead role. I enjoyed learning about your team’s priorities and discussing how to turn customer webinars into short articles that share useful insights with a wider audience.
    
    Our conversation made me even more excited about the opportunity to contribute to your marketing team. I appreciate your time and look forward to hearing about next steps.
    
    Best,  
    [Your Name]

    1 message on Pro · $0.0009 to run · 2.8 s · 96 tokens out

  • Claude Sonnet 5.5

    Passed

    Graded 4.7 of 5 on average (lowest 4).

    **Subject:** Thank you – Content Lead conversation
    
    Hi Dana,
    
    Thank you for taking the time to speak with me today about the content lead role. I enjoyed learning more about your team's goals and priorities.
    
    I was especially excited by our conversation about turning your customer webinars into short articles. It's a great way to extend the life of those insights, and I'd love to help build a repeatable process for it, from pulling key quotes to shaping each piece for search and social.
    
    Please let me know if I can provide anything else, such as writing samples or ideas for a first batch of articles. I'm looking forward to hearing about next steps.
    
    Best regards,
    [Your name]

    1 message on Pro · $0.0032 to run · 2.4 s · 212 tokens out

  • Kimi K3

    Failed

    Graded 4.3 of 5 on average (lowest 4); but 129 words, over the 120 allowed.

    Subject: Thank You – Content Lead Interview
    
    Hi Dana,
    
    Thank you for taking the time to speak with me today about the content lead role. I really enjoyed our conversation and learning more about your team's goals.
    
    I was especially excited by our discussion about repurposing customer webinars into short articles. It's a smart way to extend the life of great content, and it's exactly the kind of project I'd love to lead. I already have ideas for a repeatable workflow that could turn each webinar into a series of engaging pieces.
    
    I'm very enthusiastic about the opportunity to contribute to your team. Please don't hesitate to reach out if you need anything further from me.
    
    Thanks again, and I hope to hear from you soon.
    
    Best regards,
    [Your Name]

    1 message on Pro · $0.0018 to run · 2.6 s · 169 tokens out

10. A 30-60-90 day plan

Everyday · 4 of 4 passed

A plan you can bring to a final interview shows you've thought about the job; the limits keep it to one page.

The prompt: replace the parts in [brackets]

Write a 30-60-90 day plan for [the role]: three short lists with 3 bullets each, under 25 words a bullet.

Open in llmwise
How we filled it in and checked the replies
Write a 30-60-90 day plan for a new customer success manager at a business software company: three short lists with 3 bullets each, under 25 words a bullet.
  • Progression: Learn first, then act, then improve.
  • Realistic: Fits a new customer success manager.
  • Format: Three lists of three bullets, each under 25 words.
  • Automatic rules: exactly 9 bullet points.
  • GPT-6 Luna

    Passed

    Graded 5.0 of 5 on average (lowest 5).

    ### First 30 Days
    - Learn the product, customer segments, success metrics, and support processes.
    - Meet teammates and key customers; understand goals, challenges, and expectations.
    - Review account health, renewals, onboarding journeys, and customer feedback.
    
    ### Days 31–60
    - Own a portfolio; build relationships and document success plans.
    - Address at-risk accounts and coordinate solutions with product, sales, and support.
    - Lead customer check-ins and track adoption, outcomes, and next steps.
    
    ### Days 61–90
    - Deliver measurable progress on adoption, retention, and customer outcomes.
    - Identify expansion opportunities and share customer insights with internal teams.
    - Refine account plans and set priorities for the next quarter.

    1 message on Pro · $0.0001 to run · 2.6 s · 150 tokens out

  • GPT-6 Sol

    Passed

    Graded 4.7 of 5 on average (lowest 4).

    ### First 30 days: Learn
    - Learn the product, customer journey, success metrics, and escalation process.
    - Meet key teammates and review your assigned accounts, contracts, and support history.
    - Shadow customer calls and build relationships with your accounts.
    
    ### Days 31–60: Engage
    - Lead check-ins to understand each customer’s goals, risks, and adoption barriers.
    - Create success plans for priority accounts with measurable milestones.
    - Coordinate with support and product teams to resolve recurring issues.
    
    ### Days 61–90: Deliver
    - Run business reviews that connect product usage to customer outcomes.
    - Address at-risk accounts and identify expansion opportunities where appropriate.
    - Report progress against adoption, retention, and customer health goals.

    1 message on Pro · $0.0011 to run · 2.2 s · 149 tokens out

  • Claude Sonnet 5.5

    Passed

    Graded 4.7 of 5 on average (lowest 4).

    # 30-60-90 Day Plan: New Customer Success Manager
    
    ## Days 1-30: Learn
    - Complete product training and shadow experienced CSMs on customer calls to understand workflows, terminology, and common pain points.
    - Review your assigned accounts: contract details, usage data, renewal dates, support history, and key stakeholders.
    - Introduce yourself to each customer with a welcome call, and meet cross-functional partners in support, sales, and product.
    
    ## Days 31-60: Engage
    - Build success plans for top accounts, aligning on each customer's goals, adoption milestones, and measurable outcomes.
    - Lead customer check-ins independently, and document feedback, risks, and expansion opportunities in the CRM.
    - Identify at-risk accounts using health scores and usage trends, then launch targeted outreach to address the concerns.
    
    ## Days 61-90: Drive Results
    - Run quarterly business reviews with priority customers, presenting ROI, adoption progress, and recommendations for next steps.
    - Execute renewal and upsell strategies for accounts approaching contract dates, partnering with sales where appropriate.
    - Propose one process improvement and share customer insights with product and leadership to shape the roadmap.

    1 message on Pro · $0.0051 to run · 3.9 s · 405 tokens out

  • Kimi K3

    Passed

    Graded 4.7 of 5 on average (lowest 4).

    # 30-60-90 Day Plan: New Customer Success Manager
    
    ## First 30 Days — Learn
    - Complete product training and shadow support calls to understand common customer pain points and workflows.
    - Meet all assigned accounts; review health scores, contract details, and usage data for each customer.
    - Build relationships with sales, support, and product teams; learn escalation paths and internal processes.
    
    ## Days 31–60 — Engage
    - Conduct kickoff or check-in calls with every assigned account to establish rapport and understand their goals.
    - Identify at-risk accounts using health data; create action plans to address churn risks and adoption gaps.
    - Document customer feedback and share recurring themes with product and support teams for improvement.
    
    ## Days 61–90 — Drive Results
    - Launch success plans for top accounts with clear milestones tied to their business outcomes and renewals.
    - Identify upsell and expansion opportunities; partner with sales on qualified leads within your book.
    - Establish a regular cadence: QBRs, health reviews, and reporting to demonstrate measurable retention impact.

    1 message on Pro · $0.0025 to run · 4.3 s · 246 tokens out

Getting more from these prompts

  • Give your real facts and say "add no results I didn't mention".

  • Ask for a spoken length (about 60 seconds) rather than a word count you'll never say.

  • Practise with one question at a time, out loud.

  • Prepare your salary numbers and a fallback before the call, not during it.

How we ran and checked them

Each prompt was sent the way llmwise sends a message: the app's own system prompt, each model's own settings, and Pro's reply size limit (8,000 tokens), through OpenRouter. Every reply is shown as it came.

Rubric (graded). The grader model scores the reply from 1 to 5 on each published criterion. It passes with an average of 4 or more and no criterion under 3, and only if it also meets the prompt's automatic rules (length, words it must or mustn't use).

The grader is Claude Opus 5.5 at low reasoning effort; its own replies are graded by GPT-6 Astra, so no model grades itself. Its prompt is on our test runs page.

More tested prompts

Questions

Which AI is best for interview prep?

In our runs, GPT-6 Luna and GPT-6 Sol each passed 10 of 10, the most. The prompts with your own facts were checked above all for invented results, the thing that sinks an interview answer.

Should I memorize AI-written interview answers?

Use it to draft and to practise, then say the answers in your own words. The mock-interview prompt asks one question at a time so you can answer out loud.

Can I practise an interview here for free?

Yes. When the free-message box shows on this page, one message needs no account (GPT-6 Luna answers it); signing up gives 5 free messages, once, with no card.

Claude, GPT, Gemini, DeepSeek, Grok, Kimi, and GLM, in one chat.

See what a message costs before you send it. Free is 5 messages to try; sign in with an email link, no password or card.