AI can help you write a better CV, but only if you use it the right way. Generic tools reword what you already have. The better approach is to use AI to match your actual experience to a specific job, surface the gaps, and generate a CV that speaks to that role directly.
What most people get wrong about AI and CVs
Most job seekers open ChatGPT, paste their CV, and ask it to "make this better." The output sounds polished. It also sounds like everyone else's.
The problem is that general-purpose AI has no context. It doesn't know the job you're applying for. It doesn't know which of your achievements are most relevant. It optimizes for readability, not fit.
A CV that reads well but doesn't match the job still gets ignored. According to a 2023 report by Jobscan, over 98% of Fortune 500 companies use applicant tracking systems to filter CVs before a human sees them. If your CV isn't aligned to the job description's keywords and requirements, it won't make it through.
The solution isn't better prose. It's better targeting.
The right way to use AI for your CV
Start with your evidence, not your words
Before AI can help you, it needs raw material. That means your full work history: job titles, responsibilities, achievements, metrics, tools, and anything else that reflects what you've actually done.
Don't clean it up first. Don't try to make it sound good. Just get it all down. The AI's job is to shape that material around a specific role, not to invent things on your behalf.
Feed it the job description
This is the step most people skip. If you want a CV tailored to a specific role, the AI needs to see the job description. Not just the title. The full text.
A good AI tool will compare your experience against the requirements and tell you:
- Where you're a strong match
- Where you're weak or missing evidence
- Which of your achievements to lead with
- Which keywords to include
This is the difference between a rewrite and a tailored application.
Use AI to sharpen your bullets, not write them
AI is good at taking a vague bullet point and making it concrete. It's bad at fabricating achievements you don't have.
Give it something to work with:
"Led a project to migrate our CRM to Salesforce, took about 8 months, team of 4"
And ask it to turn that into a strong CV bullet with impact. You'll get something like:
"Led an 8-month CRM migration to Salesforce, coordinating a 4-person team and reducing manual reporting time by 40%."
The metric at the end? You'll need to supply that. AI can prompt you to add it, but it can't invent it.
Check the output against the job, not against your ego
Once you have a draft, resist the urge to judge it by how impressive it sounds. Judge it by how well it answers the job description.
Read the requirements line by line. For each one, find where your CV addresses it. If you can't, that's either a real gap or a presentation gap. Both are worth knowing before you apply.
Where general AI tools fall short
General-purpose AI tools like ChatGPT or Claude are useful for one-off rewrites. They're not built for job applications.
They have no memory of your career history between sessions. They can't score how well you fit a specific role. They don't know what ATS systems are looking for. And every output starts from scratch.
Using a general AI tool for job applications is like using a word processor to manage a spreadsheet. It can technically do the job. It's just not the right tool.
Purpose-built tools keep your experience in one place and apply it across every application. That's what makes them faster and more consistent.
What a purpose-built tool does differently
Track & Crack builds a structured profile from your CV called an applicant soul. It's a persistent record of your experience, evidence, and claims that lives between applications.
When you paste a job description, it scores your fit against the role and shows you exactly where you match and where you don't. Then it generates a tailored CV and cover letter grounded in your actual history, not a generic version of it.
The practical difference: instead of spending an hour rewriting your CV for each application, you spend a few minutes reviewing what was generated and adjusting the gaps.
If you're comparing options, this breakdown of AI CV builders covers what to look for and how the main tools differ.
A simple process that works
- Upload your full CV or paste your work history
- Add any missing achievements, metrics, or context
- Paste the job description for the role you want
- Review your fit score and gap analysis
- Generate the tailored CV and adjust anything that needs a human touch
- Repeat for each application, building on the same profile
The profile gets more useful over time. Every new role you add, every achievement you document, makes the next application faster and better matched.
FAQ
Can AI write my entire CV from scratch?
It can generate a draft, but you'll get poor results without real input. AI needs your actual experience, specific achievements, and the job you're targeting. Without that, it produces generic output that won't stand out.
Will an AI-written CV get flagged by ATS?
No. ATS systems scan for keywords and structure, not writing style. A well-targeted CV that uses the right terms from the job description will perform better with ATS, regardless of whether a human or AI helped write it.
How is this different from just using ChatGPT?
ChatGPT is a general-purpose tool. It has no memory of your career history, can't score your fit against a specific job, and doesn't know what ATS systems prioritize. Purpose-built tools are designed around the job application workflow from the start.
How many CVs should I tailor?
Every single one. Sending the same CV to ten roles is rarely better than sending a tailored CV to three. Fit matters more than volume. A CV that directly addresses the job description's requirements will outperform a polished generic one every time.