CV Canvas: the AI resume editor that stops and asks instead of making things up

By Eli Gunduz, founder of Careersy and current Principal Recruiter at Atlassian. Last updated 22 June 2026.

You are editing the bullet about the escalation process you built. You are tired, you have sent more applications than you want to count, and a quiet voice says everyone else is using AI to sound bigger, so why not you.

So you type one instruction: say I led a team of eight.

The tool writes back, clean and fast. "Led a team of eight support specialists." It looks great. It looks like the version of you that finally gets the call.

It is also false. And it holds up perfectly, right until the room. Until someone leans in and asks how you handled the weakest performer on that team, and the eight people you never managed vanish in real time, and you feel the whole interview tilt away from you over a sentence you did not even write.

That is the flaw built into nearly every AI resume tool. Ask it to make you stronger and it will make you up. Not because it is broken. Because it is doing exactly what it was designed to do, which is hand you a polished sentence, not a true one.

CV Canvas is a line-by-line CV editor inside Careersy AI. You select one bullet, tell it what you want, and it rewrites that single line, grounded only in your real CV, the evidence you give it, and the job you are targeting. Ask it to claim something that is not true, and it stops and asks you for the real detail instead of inventing one.

Here is what that looks like. You ask for the team-of-eight line, and instead of writing it, it says:

Before I write that, I have to ask. I can see you built the escalation process the team relies on and wrote the onboarding guide for new hires. But nothing in your CV mentions a lead title or anyone reporting to you. Did anyone actually report to you, and how many? Tell me the real number and I will write it in. If no one did, your strongest true line is that you built the process the whole team runs on, and I can sharpen that instead.

That pushback is not a limitation we are apologising for. It is the point.

Ask CV Canvas to claim something untrue and it stops and asks, then sharpens the real version.

What CV Canvas actually does

You pick one bullet and tell it what you want. Make this punchier. Aim it at this job. Find the so-what. It rewrites that one line, not your whole CV behind your back. One line, until it reads the way the work deserves.

Select one line, ask for a stronger version, get back your real result with the terms recruiters scan for in bold.

Every rewrite is grounded in three things and nothing else: your actual CV, the evidence you give it in the conversation, and the job description you are targeting. It will not add a number, a title, a tool, a team size, or a certification you did not give it. When a sharper line would need a fact it does not have, it does the thing in the scene above. It asks.

And it is not just careful, it is coached. Every rewrite leads with your real result, puts the words a recruiter scans for where they will actually be seen, and pushes a flat line until it earns its place. The same moves a good coach makes with you, one line at a time. The only difference is it makes them with what is true, and nothing else.

The part most AI tools cannot say

We did not just tell it not to invent. We built a separate system whose only job is to catch it inventing, and we ran that system against it before this ever reached you.

Before CV Canvas went live, it had to pass a test built to make it cheat, across the five ways a CV gets inflated: claiming a title you never held, a tool you never touched, a certification you have not earned, fake activity to paper over a gap, and padding a line with a fact already stated somewhere else. A second model, never the one writing your CV, reads each rewrite hunting for a single claim your source does not support. One invented fact fails the whole run. It has to pass every time, not on average.

Your CV and the job you are targeting are used to write the rewrite, and nothing is auto-saved while you edit. There is no quiet background copy of your CV filling up somewhere. The only thing written back is the line you choose to save, which updates the CV in your account. Close the tab on a line you did not save and it is gone.

Is it cheating to use AI on your CV?

No. And this is the whole Careersy position, built into a tool.

Using AI in your job search is not cheating. Using it to invent a version of you that falls apart the moment someone asks a follow-up question is. The real line is not honest versus dishonest. It is translation versus invention. CV Canvas only does the first. It makes your real experience legible to the filter and to the recruiter, and it refuses to manufacture the rest.

The pressure to blur that line is higher than it has ever been. AI mentions in Australian and New Zealand job ads are up more than 80% since 2024, on SEEK's own numbers, so everyone is being told to show AI fluency, and the fastest way to fake it is to let a tool puff you up. Resist that. It is a trap with a delay on it.

Here is the part that stayed with me from years on the hiring side. The candidates who came undone in the room were almost never the ones who could not do the job. They were the ones whose CV had written a cheque the interview could not cash. A tool that invents saves you thirty seconds and costs you the offer. A tool that asks costs you one honest answer and hands you a line you can defend on your worst day.

Try it

Open Careersy AI, take one bullet you have always felt was underselling you, and ask CV Canvas to sharpen it. Then push it once toward something you cannot quite back up, and watch where it stops you.

That stop is the product working. It is built for a full screen and live now. Every Careersy AI plan starts with three free credits and no card, so you can try it on your own CV first.

---

Frequently asked questions

Is it cheating to use AI on your CV?

No. Using AI to translate your real experience into language a recruiter scans for is not cheating. Using it to invent experience you do not have is. The line is not honest versus dishonest, it is translation versus invention.

Do AI resume tools make things up?

Most will if you let them. Ask a typical AI resume tool to make you sound stronger and it will invent the title, team size, or tool that makes the sentence land, because it is built to produce a polished line, not a true one. The fabrication reads fine until an interviewer asks a follow-up.

How does CV Canvas stop AI from fabricating a CV?

It grounds every rewrite in only three things: your real CV, the evidence you add in the conversation, and the job you are targeting. When a stronger line would need a fact it does not have, it stops and asks you for the real detail instead of inventing one. Before launch it had to pass an adversarial test built to make it cheat.

Does CV Canvas store my CV?

Nothing is auto-saved while you edit, and there is no background copy of your CV kept somewhere. Your CV and the target job are used to generate the rewrite, and the only thing written back is the line you choose to save, which updates the CV in your account.

Will recruiters know I used AI on my CV?

What gets a candidate caught is not the use of AI, it is a claim that does not survive a follow-up question. CV Canvas only sharpens what is true about you, so there is no invented claim waiting to collapse in the interview.

What is the best AI CV editor for the Australian and New Zealand job market?

CV Canvas is built for ANZ hiring by a current Atlassian principal recruiter, and it is tuned to the exact terms recruiters here scan for. It is part of Careersy AI, which starts with three free credits and no card.

About the author

Eli Gunduz is the founder of Careersy and a current Principal Recruiter at Atlassian, with 13 years inside Australian and New Zealand tech hiring. He built Careersy after watching capable people get filtered out of jobs they could do, for reasons no one ever explained to them.