AI for Education — Admissions Agents & Student-Services Automation

Your registrar answers maybe half the WhatsApp messages that arrive during admissions season — and a parent who waits 24 hours for a seat-availability answer has already toured two other schools. Every unanswered inquiry in January–March is a lost annual tuition contract, not a missed message. The same thirty questions eat your front office alive: fees, curriculum, bus routes, uniform, required documents, assessment dates. Staff retype the same answers hundreds of times a season while genuinely complex cases wait, and parents write in Arabic, English, and everything in between — Gulf dialect voice notes, English emails, Arabizi texts — so your Arabic-speaking desk and your English-speaking desk are never both free at the same moment. Inquiry volume spikes 5-10x in the admissions window, but headcount stays flat, so you either overstaff for eleven months or drop leads for three. Fee follow-up is manual and awkward, chased by accountants on personal WhatsApp with inconsistent reminders while receivables age. In higher education, a transcript or enrollment-certificate request takes days through a shared inbox, and advising appointments are booked by email ping-pong. Meanwhile your teachers are already using ChatGPT without policy, training, or data rules — lesson plans, student names, and internal documents pasted into consumer tools, an adoption problem and a governance problem at once. And you only find out which families aren't re-enrolling in June, too late to backfill seats.

We sell the bilingual admissions WhatsApp agent and Corporate AI Training as the two entry products — fast, low-effort, and instrumented from day one — then expand into re-enrollment campaigns, fee follow-up, and higher-ed student-services automation once the core WhatsApp infrastructure is proven.

A private K-12 school pilot (KSA/UAE, 1,000-2,500 students): a WhatsApp agent trained on the school's own fee tables, curriculum, transport zones, and document checklists, with a tour-booking flow into the school calendar and human handoff to the registrar with full conversation context — measured over 30/60/90 days by median first-response time before versus after, share of inquiries fully resolved without staff, after-hours inquiries answered, tours booked via the agent, and staff hours redirected. A university student-services pilot (5,000+ students): a web and WhatsApp desk integrated with the SIS for enrollment certificates, transcripts, and advising bookings — measured by median turnaround per request type, share of requests deflected from the human queue, and advising no-show rate before and after reminders. A school-group training pilot (3-10 campuses): a two-day applied Arabic/English workshop plus an AI-use policy template — measured by pre/post skills-assessment score delta and the number of qualified implementation opportunities the workshop surfaces. We contract to these measurable outcomes, not demos — MIT research found 95% of enterprise AI pilots produce no P&L return, and instrumentation from day one is how we stay in the other 5%. We never publish invented numbers; every pilot is measured from day one and becomes a case study only after client sign-off.

Saudi Arabia: PDPL applies to parent and student personal data, and minors' data demands guardian consent and data minimization; private schools are regulated by the Ministry of Education, with ETEC governing assessment and training standards. UAE: emirate-level regulators — KHDA in Dubai, ADEK in Abu Dhabi, SPEA in Sharjah — run inspection frameworks that reward parent-communication quality, and UAE PDPL applies alongside GDPR obligations many international schools carry via foreign parent entities. Egypt is our proving ground: Law 151/2020 applies, and the dense international-school market in Cairo and Alexandria is where we pilot cheaply before selling instrumented results into KSA and UAE at GCC pricing. Across every market we hold a strict scope line: we automate communication, scheduling, and document logistics — never admissions decisions, grading, or proctoring — which keeps us out of algorithmic-fairness controversy and high-risk AI classification.