Using AI to write your CV, without sounding like everyone else

AI is good at turning rough notes into clean CV bullets and at tailoring a CV to a job. It is bad at knowing what you actually did. Used well, it saves hours. Used lazily, it produces the same inflated CV recruiters now see every day.

Give it real material first

Before asking for anything, write down for each role:

• What you were responsible for.
• Two or three things you did that you are proud of.
• Any numbers: size of budget, team, customers, time saved, growth.
• Tools and methods you used.

The quality of the output depends almost entirely on this input.

Requests that work

• "Rewrite this bullet to start with a verb and end with the result. Keep every fact and add nothing."
• "Here is a job description and my CV. List the requirements I meet, the ones I do not, and where my CV already shows each."
• "Shorten this CV to one page. Cut the least relevant bullets for this role and tell me what you removed."
• "Suggest three ways to phrase this achievement, from plain to more assertive."

What gives an AI CV away

• Vague power words: spearheaded, leveraged, orchestrated, with no number behind them.
• Identical rhythm in every bullet.
• Skills you cannot discuss. If a tool appears on your CV, expect a question on it.
• A summary that could describe anyone.

The accuracy check

Read every line and ask: could I talk about this for two minutes in an interview? Delete or correct anything that fails. An invented number is worse than no number.

Keep it parseable

One column, standard headings, a text-based PDF. Tailor the top third for each application; that is where screening tools and recruiters look first.

PrepperApply's builder works this way: it uses your uploaded CV as the source of truth, tailors it to each job, and flags missing keywords only where they match your real experience.