How to work in artificial intelligence? Step-by-step guide: key skills, training in Morocco, online certifications, and practical advice for beginners.

How to Work in Artificial Intelligence: A Complete Beginner’s Guide

To work in artificial intelligence, you need three things: foundations in mathematics and programming, a specialization in a specific area (data, models, business applications), and concrete experience on real projects. No PhD required. What you need is a method and genuine work discipline.

Morocco is facing a serious shortage of AI experts. Companies are searching, not finding, and starting to train people themselves. That is your entry window.

Step 1: Understand What the Market Actually Wants

Before signing up for the first course you find, ask yourself one simple question: what problem do you want to solve with AI?

The market is not looking for theorists. It is looking for profiles capable of taking a business problem, identifying whether AI can address it, and deploying a solution that works in production.

The most in-demand profiles today in Morocco and Europe:

Data engineer
Data analyst
Machine learning engineer
AI integration consultant for business functions
AI governance manager

That last profile is still rare. And it is about to become critical. I cover this in my analysis of AI engineer salaries in Morocco in 2026.

Step 2: Build Technical Foundations (Without Panic)

You do not need to master everything. You need to master what is useful.

The essential foundations:

Python: the reference language for AI. Free to learn, massively documented.
Statistics and probability: not at a researcher level, but enough to understand what a model actually does.
Data manipulation: knowing how to clean, structure, and analyze a dataset.
Basic machine learning models: regression, classification, neural networks.

Where to start for free: Coursera (Andrew Ng’s machine learning course), fast.ai, Kaggle for practice. These resources are accessible from Casablanca, Rabat, Brussels, or Paris.

One rule I apply in the projects I work on: if you cannot explain what your model does to a CFO in two minutes, you have not fully understood it yourself.

Step 3: Choose the Right Training

Two main paths.

The academic path in Morocco: several universities and engineering schools offer specialized master’s programs. ENSIAS, Université Mohammed V, and UM6P in Ben Guerir have developed data and AI-oriented programs. These provide institutional credibility and a local network.

The online certification path: Google (Professional Machine Learning Engineer), Microsoft (Azure AI Engineer), IBM (AI Engineering Professional Certificate on Coursera). These certifications are recognized by international recruiters and can be prepared alongside a job.

My advice: combine both if you can. A local master’s degree plus an international certification is a profile that is hard to overlook.

I detailed the best options in my complete guide to AI training in 2026. Read it before committing financially.

I have built a diagnostic framework to help executives identify the AI profiles they actually need in their organization. Download the Board Pack AI 2026 if you are on the employer side and want to structure your recruitment.

Step 4: Build a Portfolio, Not a CV

This is where most candidates go wrong.

An AI recruiter does not want to read that you completed 12 online courses. They want to see what you built.

What your portfolio must contain:

Two or three projects on GitHub with clean, documented code
A project addressing a local issue (Moroccan data, a real sector problem)
A demonstration of your ability to communicate results, not just produce code

Kaggle is your training ground. Enter competitions, even if you do not win. Your ranking is visible. Recruiters look.

Step 5: Enter the Market Through the Right Door

Big tech companies are not the only entry point. Often, they are not the best one for a beginner.

Realistic entry points:

Consulting and integration firms deploying AI solutions for their clients
Moroccan and African AI startups looking for versatile profiles
Large Moroccan companies (banks, telecoms, retail) internalizing their data teams
International operators setting up in Morocco, like Orange with its recently deployed sovereign generative AI solution

Networking matters as much as skills. Attend tech events in Casablanca and Rabat, and follow AI communities online. Opportunities circulate before they are posted.

Mistakes to Avoid

First mistake: accumulating certifications without ever building anything. Certifications open doors. They do not replace experience.

Second mistake: aiming too broadly. “I want to do AI” is not a strategy. “I want to analyze customer data in the Moroccan banking sector” is.

Third mistake: ignoring non-technical skills. The ability to explain a model to a board, manage a project with multiple stakeholders, understand regulatory compliance issues: these are the skills that separate a good technician from a profile that advances.

Fourth mistake: waiting until you are ready. You will never be 100% ready. Start with what you have.

What You Can Expect

The AI market in Morocco is under pressure. Companies struggle to recruit qualified profiles, and this shortage is well documented. A serious profile with solid foundations and a real portfolio finds opportunities.

The world’s leading AI companies are also recruiting profiles based in North Africa for remote or hybrid positions. This is a 2026 reality that many Moroccan candidates are not yet leveraging.

The question is not whether you can work in AI. The question is whether you are ready to invest 12 to 18 months of serious work to get there.

If yes, the market is waiting for you.

If you are a CHRO or CEO looking to structure your AI recruitment strategy, request a free diagnostic. I work with organizations in Morocco, Belgium, and France on exactly this topic.

FAQ

Do you need a computer science degree to work in AI?

No. Profiles from mathematics, physics, economics, and even the social sciences have successfully transitioned into AI. What matters is the ability to learn, reason about data, and solve concrete problems. A computer science degree helps, but it is not a prerequisite.

How long does it take to become operational?

With regular, structured work, a profile can reach a junior operational level in 12 to 18 months. This requires serious training, daily practice, and concrete projects. There is no credible shortcut below that timeframe.

What are the best free resources to get started?

Coursera (Andrew Ng’s machine learning and deep learning courses), fast.ai for hands-on learning, Kaggle for data and competitions, and the official documentation for Python and libraries like scikit-learn and TensorFlow. These resources are accessible from anywhere.

Does the Moroccan market actually hire AI profiles?

Yes. Banks, telecoms, consulting firms, and tech startups are actively recruiting. The shortage of qualified profiles is real. International companies setting up in Morocco are amplifying this demand. It is a favorable moment to enter this market.

Working in AI: The Beginner's Complete Guide