What does an AI trainer actually do?

Takes the tools everyone is talking about and makes one specific team able to use them in their daily work. Not a lecture about the future of technology, a session where people work on their own real documents and tasks until the way they work changes.

How do we tell whether a trainer is any good?

Ask them about the last time an adoption failed at a client, and why. Someone who has actually worked inside organisations will have a specific answer. Someone who mostly gives presentations will speak in generalities.

Is the training in Arabic or English?

Either, at the same standard, with no interpreter. An Arabic session is not a translated English one. Many teams choose a mixed format: explanation in Arabic, technical terms in English.

How long is a session?

From half a day for the executive briefing to two days for Python and data fundamentals. Most programmes are a full day, which is long enough for real hands-on work without exhausting people.

What is the right group size?

Between 8 and 20. The training is hands-on and everyone works on their own tasks, which needs enough attention per person to be useful. Larger groups are possible but the session becomes a lecture.

What happens after the session?

The team leaves with the materials, the prompts each person built for their own work, and whatever they actually produced during the training. A short follow-up some weeks later is available for organisations that want to measure what stuck.

Do you cover data confidentiality requirements?

Yes, and as part of the curriculum rather than an addendum. Sessions cover what happens to text pasted into consumer tools, the difference enterprise accounts make, and how to read your own organisation's policy against what the tools actually do.

With a free 15-minute call to understand what the team does and what you want to change. If training is not the right answer, I will say so.

United Arab Emirates

AI trainer in the UAE

I train government entities, corporates and universities across the Emirates to use AI in the work they already do, in Arabic and in English, at the same standard in both.

Book a free 15-minute call See training programmes

Most organisations in the Emirates have already bought the tools. Licences are assigned, a policy has been drafted, and somewhere there is a slide deck about transformation. What is usually missing is the part where people actually change how they work on a Tuesday afternoon. That gap is what an AI trainer exists to close, and it is the only thing I am measured on.

I deliver corporate and government AI training across the UAE, in Arabic and in English. Before AI I spent years as a journalist, a cinematographer and a creative director, and I co-founded Python Arabic Community, the Arab world's largest AI and Python education channel. Both of those matter more than they sound like they should, and I will come back to why.

What an AI trainer actually does

The job is not to explain what a large language model is. Almost nobody in a training room needs that, and the ones who think they do usually need something else.

The job is to take a specific team, their documents, their approval chains, their reporting cycle, the software they are already required to use, and make them genuinely faster at the work in front of them. That means:

Working on real files brought to the session, not sanitised examples.
Building a routine for checking output, so nothing unverified reaches a colleague or a client.
Being explicit about what must never be pasted into a consumer tool, and why.
Leaving each person with prompts built for their own recurring tasks.

A good session produces something. An automation that runs. A script that works on their data. A prompt library the team shares. If everyone leaves with only notes, the training has failed, whatever the feedback form says.

How UAE organisations are actually using AI training

The pattern here is distinctive, and it is worth being specific about rather than reaching for global statistics.

Government entities tend to start from governance. The question is rarely "can this write our reports" and almost always "what happens to the data, who is accountable for the output, and what do we tell the auditor". Sessions for federal and local entities spend real time on data residency, on the difference between enterprise and consumer accounts, and on reading their own policy against what the tools do in practice. The training only works if it survives contact with a compliance officer.

Corporates usually arrive after a pilot stalled. Someone enthusiastic ran a trial, it went well, and then nothing spread. That is almost never a technology problem, it is that the pilot was run by people who would have found a way regardless, and the rest of the team had no reason to change. Training that works targets the unenthusiastic majority and their actual weekly tasks.

Universities are split between two very different needs: helping academic and administrative staff use the tools, and working out an institutional position on students using them. The second is a policy conversation, not a training one, but it comes up in every session and pretending otherwise wastes everyone's time.

What makes an AI trainer effective in this market

Three things separate training that changes behaviour from training that fills a day.

Delivering natively in Arabic

Most trainers in this market deliver in English only. For a lot of teams in the Emirates that is fine. For plenty of others it quietly halves what the session achieves, people follow the words but stop asking questions, and the questions are where the learning happens.

I deliver the same session in either language. Not a translated version, and not with an interpreter sitting between me and the room. Many teams choose a mixed format: explanation in Arabic, technical terms left in English, because that is how the team actually talks to each other. Latin terms stay Latin. ChatGPT is ChatGPT, not a transliteration nobody uses.

Teaching what you build

I publish technical breakdowns of frontier AI systems to the Arab world's largest AI and Python education channel. That is not a credential to put on a slide; it is a testing mechanism. Every explanation I use in a training room has already been tried on tens of thousands of people, and the ones that did not land were rewritten. You are getting the version that survived.

A background in making hard things clear

Before AI I worked as a journalist, a cinematographer and a creative director. A journalist's entire craft is taking something complicated and making it clear to someone who is not an expert and did not ask to become one. That is the same job as corporate AI training, performed under the same constraint: if the audience does not follow, it is your failure, not theirs.

The programmes

Six programmes, each built around what a specific kind of team needs. They can be delivered as they are, or combined and adjusted for one organisation.

Generative AI for the workplace: a full day on ChatGPT, Claude, Gemini and Copilot applied to work the team already does.
Prompt engineering that works: reliable, repeatable results from real documents.
AI agents and workflow automation: one live automation built during the session.
AI for content, media and design teams: a faster pipeline that still looks like your brand.
Python and data fundamentals for non-engineers: enough independence to stop queueing for a developer.
Executive AI briefing: half a day on what to adopt, what to ignore, and where the risk sits.

Formats and delivery

On-site anywhere in the Emirates, or online. Most government entities prefer on-site; distributed teams usually prefer online. Sessions run from half a day to two days, with a full day being the most common, long enough for hands-on work, short enough that people are still thinking clearly at the end.

Group size between 8 and 20 works best. The training is hands-on and everyone works on their own tasks, which requires enough attention per person to be useful. Larger groups are possible, but past about 25 the session becomes a lecture and the outcomes change accordingly. I will tell you that before you book rather than after.

Who I work with

Federal and local government entities, corporates, and universities across the UAE. Some engagements are covered by confidentiality terms, so they appear on the sector rather than by name, and I will name them on a call once permission is in place. Where a client is happy to be named, they are.

How an engagement works

A free 15-minute call. What the team does, what you want to be different afterwards, and whether training is even the right instrument. Sometimes it is not, and I will say so.
A short written proposal. Which programme, which language, how long, what the team leaves with. One page, not a deck.
Preparation. I ask for real examples of the work, anonymised where needed, so the session runs on your material rather than on generic samples.
The session. Hands-on throughout. Everyone works on their own tasks.
What you keep. Materials, each person's prompt set, and whatever was built during the day.
Optional follow-up. A short session some weeks later to see what stuck and fix what did not. This is where you find out whether the training actually worked.

If that sounds like what your team needs, book the 15-minute call and we will find out quickly.

Common questions

: Takes the tools everyone is talking about and makes one specific team able to use them in their daily work. Not a lecture about the future of technology, a session where people work on their own real documents and tasks until the way they work changes.
: Ask them about the last time an adoption failed at a client, and why. Someone who has actually worked inside organisations will have a specific answer. Someone who mostly gives presentations will speak in generalities.
: Either, at the same standard, with no interpreter. An Arabic session is not a translated English one. Many teams choose a mixed format: explanation in Arabic, technical terms in English.
: From half a day for the executive briefing to two days for Python and data fundamentals. Most programmes are a full day, which is long enough for real hands-on work without exhausting people.
: Between 8 and 20. The training is hands-on and everyone works on their own tasks, which needs enough attention per person to be useful. Larger groups are possible but the session becomes a lecture.
: The team leaves with the materials, the prompts each person built for their own work, and whatever they actually produced during the training. A short follow-up some weeks later is available for organisations that want to measure what stuck.
: Yes, and as part of the curriculum rather than an addendum. Sessions cover what happens to text pasted into consumer tools, the difference enterprise accounts make, and how to read your own organisation's policy against what the tools actually do.
: With a free 15-minute call to understand what the team does and what you want to change. If training is not the right answer, I will say so.

Start with a short call

Fifteen minutes to understand what your team does and what you want to change. If training is not the right answer, I will say so.

Book a free 15-minute call Get in touch