CV with AI Mozambique: What to Automate and Avoid

Creating a CV with AI in Mozambique is fast — the risk lies in what comes next. The machine writes text that is too clean, lacking personality; it smells robotic, and that smell is what sinks your profile during screening. Used wisely, AI takes away the tedious part. Used carelessly, it is the perfect excuse to lose the job.

In the Inademy Guide, we show you where the machine fits in, where it hurts your chances, and how to make your CV work on both layers: the filter and the recruiter.

The short answer

• What you can automate: the first draft of your profile, tailoring to each job ad, checking keywords, and formatting.
• What you never delegate: your numbers, real-world context, and deciding what to include.
• Two readers: your CV has a machine reader (the filter) and a human reader (the recruiter) — the same document has to convince both.
• The main trap: over-optimising. Keyword stuffing and text that sounds like AI will get you rejected at the second stage, even if you pass the first.

Why your CV now has two readers

Applications now go through automated screening systems, known as ATS: software that reads your CV, extracts the text, and ranks candidates against the job posting. This process happens before any human touches your document.

In a 2024 global report by LinkedIn, over 90% of recruiters use these systems when receiving high volumes of applications for a role. In practice, this changes everything: before anyone reads your CV, a machine has already ranked you.

That is the shift. The old CV only had to impress one person. Today's CV must pass through two layers: the first extracts and ranks (the machine), while the second evaluates (the recruiter).

A CV written by AI without your own review usually passes the first layer and fails at the second—because the machine won't invent metrics you don't have, and a human instantly notices when the writing isn't yours.

What you can safely automate

This isn't about banning AI. It's about using it where it excels and stopping it where it makes things up. Here are four tasks where AI assistance makes sense:

• First draft of your profile summary. You provide the facts — years of experience, roles, industry — and ask for a draft of your professional summary. You write the final version yourself, in your own words.
• Tailoring to each job ad. For every application, ask the AI for a list of key terms from the job posting and check which ones are missing from your CV. This is the most repetitive task of all — perfect for automation.
• Formatting check. A clean layout, sections in the right order, and a single readable file. The machine spots formatting issues human eyes easily miss.
• Final audit. Before sending, ask for a quick check: which words from the job post are missing, which dates don't line up, or which statements lack supporting examples.

The rule is simple: AI gets the work started, but you decide the content. What comes out of the machine is a starting point — never your final document.

Where AI costs you the job

Here is the heart of the matter. Six situations where an AI-generated CV, created without care, holds back your application:

• Keyword stuffing. Repeating the same skill ten times works worse than it seems: modern filters do not just count words, and a recruiter reads an overstuffed CV like a promotional ad.
• The smell of AI. Uniform sentences, stacked adjectives — "highly motivated", "results-oriented", "excellent skills" — repeated from job ad to job ad. Anyone receiving fifty CVs a day recognizes the pattern in seconds.
• Lack of numbers. AI writes "responsible for the team" — generic, proves nothing. The recruiter wants to see "reduced delays by 20% in six months". Only you know your numbers.
• The "creative" format that breaks the parser. Two columns, graphics, photos in odd places: the system cannot read it, and your CV reaches the filter half-empty.
• Made-up skills. Asking AI to "complete" your CV with certifications you do not have. It passes the filter, but falls apart in the interview — and lost trust cannot be recovered.
• One single CV for fifty job ads. The opposite mistake: using AI once a year and applying for everything with the exact same document. The value of AI comes per application, not per file.

The complete list of errors that kill your CV is in our own guide, with examples of what to put instead (the errors that eliminate your CV).

Three signs your CV smells like a robot

• The same "good" phrases appear in all your applications.
• Lots of adjectives, almost no facts.
• No reference to the local market: sector, Mozambican context, real working conditions.

If two of these signs apply to you, your CV no longer represents you — it represents the language model that wrote it.

The filters you don't even need AI to pass

Good news: most filtering rules are mechanical — and have always been checked manually. Five minutes of manual work separate a CV that gets read from one that gets ignored:

• A single file, in PDF format, unless the job ad asks for Word.
• Sections in standard order: experience, education, skills.
• The exact words from the job posting, in the order they appear.
• Real, readable, up-to-date contact information.
• Zero columns, tables, or graphics.

If you want to see the full step-by-step mechanics, Inademy's practical ATS guide covers exactly this — including how to test whether your file is readable by a system.

The 7-second test

After the filter, there's a human — and they're in a hurry. An eye-tracking study by Ladders, conducted in 2012 and repeated in 2018, measured recruiters' first contact with a CV: the initial scan lasted, on average, between 6 and 7.4 seconds per page.

In those seconds, the eye goes straight to the top: your profile, your most recent experience, and about three concrete results. A three-page CV packed with "perfect" AI-written paragraphs fails this test — the best information, buried at the end, simply doesn't get read.

Try the test at home. Read the first page out loud in thirty seconds, as if you had to explain it. Anything you can't defend out loud, remove or rewrite.

The myth of the "75% rejected"

There is a number floating around: 75% of CVs die in the filter before a human ever sees them. It is the most repeated figure on the subject — and also the one without any real study behind it. Its origin traces back to 2012 marketing material from an American company that closed in 2013, with no public methodology.

Approach this with common sense. There is no need to panic — though it isn't time to relax either. Filters do exist and rank applications; what isn't true is that "everyone gets rejected," and that very idea is what leads people to make the bad choices on this list.

If you see this number repeated elsewhere, be skeptical. Recent independent reviews have already pointed out that the statistic has no foundation — reality is calmer than the panic, but it requires the exact same effort.

How to use AI to your advantage, step by step

Thirty minutes per application, split like this:

1. Copy the job ad into a notepad.
2. Ask the AI for a list of five to ten exact keywords from the ad.
3. Manually check which ones you already have in your CV and in what context.
4. Generate a profile draft using your numbers — and rewrite everything in your own words.
5. Do the 7-second test on the first page.
6. Save the CV with the name of the job ad. Never use a "generic" file for everything.

Five minutes of machine work, twenty-five of yours. If you flipped that ratio, the job ad would be writing your CV — and the result would have the exact robotic feel we described above.

What about jobs in English?

For NGOs, multinationals, and international roles, your CV needs to be in English. The rules don't change — only the language. The trap here is asking AI to "translate" your Portuguese CV: it gives you generic English that leaves out your local achievements.

The best approach is to start with your Portuguese CV, which already has your facts and figures, and then adapt the language. Inademy's English CV guide has sample phrases and the right structure for the international market.

How Inademy helps you

You can build your CV and apply in the same session. Inademy offers a complete CV guide and free Word CV templates for you to download before you start.

You apply through the job board: create a free profile, upload your CV in PDF format, and receive alerts by email or WhatsApp whenever a new position opens in your sector.

One rule keeps everything honest: AI gets things started, you decide. The tools serve your story—they never replace it.

Where these numbers come from

• Ladders, eye-tracking study (2012, repeated in 2018): 6 seconds and 7.4 seconds on the initial review of a single-page CV — cited by specialized publications.
• LinkedIn, Global Talent Trends (2024): over 90% of recruiters use screening systems for high-volume vacancies.
• The "75% of CVs rejected" statistic: originated in marketing material from 2012-2013 with no public methodology; recent independent checks show it has no basis in fact.
• This article does not include amounts in meticais — no salaries are involved.

Frequently asked questions

Should I just let AI write my whole CV and be done with it?

You can — but it's worse than not using it at all. The text comes out polished yet generic, with that robotic feel mentioned above. The right way to use AI is for drafting, checking, and then rewriting it yourself.

Is a CV photo a problem for screening filters?

Photos are still the norm in the Mozambican job market, so having one won't hold your application back. What actually blocks you are complex layouts that systems can't read. For international roles, the opposite applies — omit the photo there.

Will using AI to build a CV in Mozambique cost me more job opportunities than it wins?

It depends on how you use it. If you stuff your CV with buzzwords and send the same document everywhere, you will lose out. If you tailor each application and check for missing details, you will save time and improve your fit — without making anything up.

Should I tell the recruiter I used AI?

You don't have to. What matters is that you can back up everything written on the page: every number, every result, every skill. If your text is full of phrases you can't explain in an interview, the recruiter will notice — confession or not.

Should I send my CV as a PDF or Word document?

Send a PDF unless the job post asks otherwise. A PDF preserves your layout, while Word makes editing easier. Just make sure the file contains readable text — most systems cannot read a scanned image.

What if I hear nothing back from the filter?

Re-read the job ad and align your exact wording. Check your section order and test the file: copy the text — if nothing pastes, the format is broken. If you still hear nothing, try applying by email with a short subject line and body text; our guide on sending a CV by email includes a template.

In summary, what to do today

1. Write a core version of your CV using facts and numbers, starting from a free Word template.
2. Pick three real job ads and copy the exact words from each one.
3. Ask AI to draft a profile summary for each ad — then rewrite everything in your own words.
4. Do the 7-second test on the first page of each CV.
5. Apply for at least three jobs today on the Inademy job board.