AI For Students runs in schools and on campus. Students take a real problem, build an AI solution to it, and put it in front of people who ask hard questions — investors in a school, a paying client at university. Below: the pilot that started it at Kaumeya Language Schools with Edge, the university programme at MSA and ESLSCA, and the skills roadmap behind both.
Kaumeya Language Schools · Alexandria
×
This is the part schools ask about most, so we will be honest about it. The first hour was flat. Fifty teenagers in a room, arms folded, no idea what any of this had to do with them. What changed it was not a better slide deck — it was handing them a problem they actually cared about and the tools to solve it before the bell went.
No brief they believed in. AI as a thing that happens to other people, in other countries, for other jobs. The usual classroom posture: wait for instructions, do the minimum, go home.
Every student in a team. Every team with a product, a number and a story. Students staying past the session to finish a build, arguing about pricing, rehearsing a pitch in the corridor.
No lecture series, no certificate at the end of a video course. Four working sessions on campus, each one ending with something that did not exist that morning.
Teams form and go hunting for something genuinely broken in their own school — not a hypothetical, something they can point at.
From prompt to product. Teams learn to direct AI models properly — not to ask for answers, but to specify, iterate and ship.
An idea that cannot pay for itself is a hobby. Teams size the market, price the product and build a financial model they can defend.
A real panel, real questions, no soft landing. Each team gets the floor, the demo runs live, and the numbers get challenged.
Teenagers do not have an information problem. Everything we could teach them is already a search away. What they have is an attention problem — so the entire programme is built as a game with real stakes, and almost none of it is us talking.
Ten teams of five, each one a company with a name and roles. You are not a student in row three any more — you are the person your team is waiting on.
Scores go to teams that produce, not teams that plan. Every round ends with something on a screen, and the board updates in front of everyone.
Sprints measured in minutes, not weeks. Deadlines close fast enough that overthinking is not an option and the only way through is to build.
No tool is introduced in the abstract. Every technique arrives at the moment a team needs it to get unstuck, which is the only moment it sticks.
Teams show their work to the room, not to a marking sheet. Peer pressure turns out to be a far better motivator than a grade.
An investor panel and a winner. The competition is what makes the last two weeks feel like something worth staying late for.
Nothing in this programme is a teaching sandbox. Students work in the production tools, and they learn the advanced end of them — how to brief a model, how to iterate on its output, how to get from a paragraph of intent to a deployed product.
The thinking partner. Problem framing, user interviews, product specs, business logic and the writing that surrounds all of it.
Ideation and fast iteration. Where teams stress-test a concept, generate options and argue with a second opinion before committing.
Where the code lives and runs. Teams go from idea to a hosted, working application without ever setting up a laptop environment.
Interface and product surface. A team describes the app it wants and gets something real enough to put in front of a user the same afternoon.
Generative video and imagery. The brand, the demo film and the pitch visuals — made by the team, not bought from a template.
A team only counted as finished when it had all three. Ten teams made it to the panel, one of the ten is now running inside Kaumeya as a real school system — and five of the fifty students walked out of the programme with an internship.
Not a mockup and not a slide. A functioning AI solution to a problem the team found on their own campus, built and hosted by the students themselves, and tested on the classmates it was meant to serve.
Market size, pricing, cost base, and a projection they had to justify line by line. The moment a student has to defend their own assumptions is the moment the idea stops being a school project.
Story, live demo, numbers and an ask — delivered to a panel that asked the questions investors actually ask. Sixteen-year-olds fielding challenges on retention and unit economics, in front of the whole cohort.
The pitch panel is not a performance. People in that room were looking for talent, and five of the fifty were offered real positions off the back of what they built — before they had finished school, and without a CV between them. That is the whole argument for teaching this way: the work is the credential.
At university level the brief changes. Students are already studying data, AI and engineering — what they are missing is a real client, a real dataset and someone senior watching the work. So we bring all three. We upskill them, mentor them through the build, and pair every team with an actual company that needs the solution. Their faculty supervise the academic side; our practitioners supervise the engineering.
Data & AI engineering students
Each hackathon takes a company with a genuine problem and a genuine dataset, and puts a team of Data and AI students on it. Not a case study, not a sanitised sample set — the client's own data, the client's own constraints, and a solution they can use.
An AI solution built for one of the most-watched business formats in the country — applied to how the show finds, sorts and tells its stories.
Arabic-language business media at scale. The team worked the language problem head on — the part most off-the-shelf tools quietly fail at.
Real clinical operations data across five governorates. Multi-site, messy and sensitive — which is exactly the condition healthcare AI has to work in.
In progress
Twelve students from the College of Engineering, working inside Egypt Eternal Voices — our national generative-AI storytelling programme run with Dr. Zahi Hawass. Their brief is to make films about Egypt's ancient history, pairing verified historical research with generative video. Engineers learning to direct a story is not a detour from technical work. It is the fastest way we have found to teach judgement about what these models actually produce.
A single focused group rather than a mass intake — small enough that every student takes a real role on a real film.
Historical accuracy first, generation second. Research the period, then use the tools to show it — not the other way round.
The films come out of the edit and onto a screen, in front of a jury and an audience — the same standard as every other Eternal Voices entry.
We do not teach to a syllabus, we teach to a roadmap. Each path names the skills a student walks away holding, in the order they pick them up, and ends in a piece of evidence — a shipped product, a client reference, a finished film — that an employer can look at. That is what turns a course into an internship.
A product other students actually use, a financial model, and a recorded pitch to a panel. At Kaumeya, five of the fifty were hired into internships off the back of it.
A solution built on a real company's data, delivered to that company. Shark Tank Egypt, Business Bel3arby and BeWell Clinics are on these students' CVs as clients, not as coursework.
A complete short film made with generative tools, screened and judged. Egypt Eternal Voices runs with Dr. Zahi Hawass.
Alexandria. Fifty students, ten teams, ten solutions built and pitched, one now implemented and in daily use inside the school, and five students into internships. Delivered with Edge.
The second school cohort is under way at Dar El-Tarbia in Zamalek. On campus, MSA students are shipping to three real clients and twelve ESLSCA engineers are making films inside Egypt Eternal Voices.
Not a slogan. A test we hold every programme to — if it only reaches Cairo, or only reaches people who already had a head start, we haven't done the work.
Get into the rooms where this actually happens — meetups, GenAI Nights, hackathons and the festival.
See the benefits → For organisationsPut your organisation behind the national target. Sponsorship, training or an innovation challenge.
See the options →