These ChatGPT resume prompts help you write a stronger resume from your real experience: they pull out accomplishments you forgot, match your wording to a job posting, fix weak bullets and check the result. Every prompt tells the model to use only facts you supply and to mark gaps instead of filling them, because a resume line you can't defend in an interview is worse than no line at all.
How to use these prompts
Copy a prompt into ChatGPT, Claude or Gemini and replace the blanks in double braces, or keep the default after the colon: {{role:product designer}} means "product designer" unless you change it.
Start with the raw material prompts (1 to 3) to build a plain-text list of what you have actually done. That list is your source of truth for every other prompt. Paste it in each time, along with the job posting when you're tailoring.
Before pasting, remove your phone number, address and email. The model doesn't need them to improve a bullet point.
How do you keep ChatGPT from inventing experience?
Language models fill gaps with plausible text. On a resume, plausible text is a liability: an invented metric or a skill you've never used will come up in an interview or a reference check. Anthropic's guide to reducing hallucinations recommends explicitly giving the model permission to say it doesn't know, and restricting it to information from the documents you provide. Every prompt here does both with the same rule:
Use only facts from my notes. If something would make a bullet stronger
but isn't in my notes, write [ADD: what's missing] instead of inventing it.
When you see an [ADD: ...] placeholder, either fill it with a real fact or delete the claim. Never let a placeholder become a guess.
Which prompts build your raw material?
1. Interview me about my work
Most people undersell themselves because they can't remember what they did. Questions one at a time work better than a blank page.
I'm updating my resume. I work as a {{role:product designer}} and want to
move toward {{target:senior product designer roles}}.
Interview me about my current and recent jobs to find accomplishments worth
putting on a resume. Ask one question at a time and wait for my answer.
Ask about problems I solved, things I built or improved, people I helped,
results I can measure, and anything I was trusted with. Follow up when an
answer is vague ("How many?", "What changed afterward?").
After {{questions:12}} questions, list the accomplishments you heard, in my
words, with no embellishment.
2. Turn notes into resume bullets
Paste a brag document, performance review notes or a messy list. The output is bullets you can trust because they only restate your notes.
Turn my notes below into resume bullets for my time as {{role:product designer}}
at {{company:my current company}}.
Rules:
- Use only facts from my notes. Where a number or result would help but
isn't there, write [ADD: what's missing].
- Start each bullet with a plain past-tense verb (built, cut, led, wrote).
- Show the action and the result in one line where the notes allow it.
- No buzzwords: avoid "spearheaded", "synergy", "dynamic", "results-driven".
- Give {{count:6}} bullets, strongest first.
Notes:
{{notes}}
3. Find numbers I already have
You often have more numbers than you think. This prompt asks where to find them instead of making them up.
Here are my resume bullets. For each one that has no number, ask me a
specific question that could lead to a real one: team size, users, money,
time saved, volume, frequency, or a before and after. Suggest where I might
find the answer (dashboards, old emails, performance reviews, tickets).
Do not suggest any number yourself.
{{bullets}}
Which prompts tailor your resume to a job?
4. Match my experience to a job posting
Do this before rewriting anything. It shows what to lead with and where the honest gaps are.
Compare my experience to this job posting.
1. List the posting's requirements, split into must-haves and nice-to-haves.
2. For each one, quote the line from my resume or notes that shows it, or
write "No evidence" if nothing does. Don't stretch a loose match.
3. Rank my three strongest matches.
4. For each "No evidence" item, tell me if I should leave it alone, mention
related experience honestly, or address it in a cover letter.
Job posting:
{{posting}}
My resume or notes:
{{resume}}
5. Tailor bullets without adding claims
Uses the posting's words only where they truthfully describe what you did.
Rewrite my bullets below for this job posting. You may reorder them, change
emphasis, and use the posting's terms where they accurately describe my
work. You may not add skills, tools, responsibilities or results that aren't
in my bullets. After the rewrite, list each change and the original line it
came from.
Job posting:
{{posting}}
My bullets:
{{bullets}}
6. Write a professional summary
Two or three lines at the top. Most summaries say nothing; this one has to point to evidence.
Write three versions of a two or three line resume summary for
{{target:senior product designer roles}}, based only on the resume below.
Each line must point to something specific in the resume (years, domain,
type of work, a result). No adjectives about my personality like
"passionate" or "detail-oriented". Mark any claim that the resume doesn't
support.
Resume:
{{resume}}
7. Build a skills section from evidence
A skills list should only hold things you could discuss in an interview.
From the resume below, build a skills section for {{target:senior product designer roles}}.
Only include a skill if a bullet shows me using it. Group them (for example
tools, methods, domains). For each skill, note the bullet that supports it.
Then list skills from this job posting that I left out and that my resume
does show, and skills it asks for that I have no evidence of.
Resume:
{{resume}}
Job posting (optional):
{{posting}}
Which prompts fix weak spots?
8. Rewrite duty bullets as results
"Responsible for X" says what the job was, not what you did in it.
These bullets describe duties, not accomplishments. For each one, ask me
one question that would reveal the result or impact. After I answer, rewrite
the bullet using my answer. If I can't answer, keep the bullet short and
factual rather than inflating it.
{{bullets}}
9. Cut it to one page
For when the resume has grown over the years. Asking for reasons makes the cuts easy to review.
My resume is {{current_length:two pages}} and I need it on one page for
{{target:senior product designer roles}}.
Tell me what to cut or shorten, in order of least loss first: old roles to
compress, bullets that repeat the same skill, details that don't matter for
the target role. Give a reason for each cut. Then show the shortened
version. Don't rewrite bullets that you keep, apart from shortening them.
{{resume}}
10. Explain a gap or a career change
Honest framing works better than hiding it. This prompt keeps the explanation short and true.
I need to address {{situation:a 14-month gap after a layoff}} on my resume
and in interviews. Here's what actually happened: {{facts}}
Give me:
1. One line for the resume (or advice to leave it off the resume and cover
it elsewhere).
2. A two-sentence version for a cover letter.
3. A spoken answer for an interview, under 30 seconds.
All three must be truthful and based only on what I told you. Keep the
focus on what I'm ready to do next.
11. Skim test as a recruiter
A recruiter often skims before reading. This prompt shows what comes across in that skim.
Read my resume the way a busy recruiter for {{target:senior product designer roles}}
would on a first skim. Then tell me:
1. What you remember: role, level, strongest result.
2. What you'd want to know that the resume doesn't make easy to find.
3. Any line that sounds generic or that you'd doubt.
4. Formatting problems: dense blocks, inconsistent dates, unclear titles.
Give the five fixes that would matter most, without changing any facts.
{{resume}}
Which prompts help with the rest of the application?
12. Cover letter from resume and posting
Short, specific and built from what you already wrote.
Write a cover letter for this job, under {{words:250}} words, using only
facts from my resume. Open with why this role at this company, using
something specific from the posting. Then two short paragraphs, each tying
one of my real accomplishments to a need in the posting. End with a plain
closing line. No clichés like "I am writing to express my interest".
Job posting:
{{posting}}
Resume:
{{resume}}
13. LinkedIn headline and About section
Keeps your profile consistent with the resume, so the two never contradict each other.
From my resume below, write:
1. Five short LinkedIn headline options, specific to
{{target:senior product designer roles}}.
2. An About section of 150 to 250 words in first person, in a plain and
friendly voice, using only facts from the resume.
Every claim must match the resume exactly; don't round numbers up or
upgrade titles.
{{resume}}
14. Final fact check before sending
Run this last, every time. It catches anything that drifted during rewrites. Anthropic's hallucination guide suggests the same move: ask the model to back each claim with a direct quote and retract any claim it can't support.
Compare my final resume against my source notes. For every line on the
resume:
- Quote the part of my notes that supports it.
- If nothing supports it, or the resume overstates it (bigger number,
higher title, "led" where the notes say "helped"), flag it.
Give a table: resume line, supporting note, status (supported, overstated,
unsupported). Don't fix anything; just report.
Final resume:
{{resume}}
Source notes:
{{notes}}
What should you check before you send it?
Read every line out loud and ask: could I talk about this for two minutes in an interview? If not, cut it or change it. Check that dates, titles and employer names match your LinkedIn profile and any background check paperwork. Make sure no [ADD: ...] placeholders are left.
Mind your data, too. Remove contact details before pasting and check your app's privacy settings. In Claude's consumer plans (Free, Pro and Max), the Model Improvement setting described in Anthropic's model training FAQ controls whether your chats can be used to improve Claude.
To prepare stories for the interviews that follow, see the STAR story and behavioral round prompts in technical interview prompts. OpenAI's ChatGPT prompting guide also recommends iterating: "Try something, fail, learn from the failure, and build a better prompt."
Keep your job search prompts ready
A job search means running the tailoring prompts again for every posting. Kipt keeps prompts like these as Markdown in a private GitHub repo you own, with blanks such as {{target}} and {{posting}} ready to fill, and ⌘K to find the right one for each application.