The recently completed AI4EAC Innovation Challenge showed that Africa’s next generation of AI innovators is not waiting for the future to materialize – they are already leading the way, using their AI and coding skills to solve local problems and shape a more inclusive tomorrow.
This six-month regional AI skills and innovation programme combined AI training, mentorship and webinars with practical challenges in education, agriculture, health and finance, culminating in a two-day innovation challenge held in March 2026.
Organized under the East African Community (EAC) AI Alliance, and implemented with partners including UNESCO Campus Africa, Germany, the Zindi network of African data scientists, the Inter-University Council for East Africa (IUCEA), the East African Science and Technology Commission (EASTECO), and the Japan International Cooperation Agency (JICA), the challenge drew close to 1,000 students from 57 universities across eight countries: Burundi, Democratic Republic of the Congo, Kenya, Rwanda, Somalia, South Sudan, the United Republic of Tanzania and Uganda.
Harnessing AI to fill labour market gaps
Among the challenge categories, the Skills2Job education track challenge, led by UNESCO, explored one of the region’s most urgent questions: how to better connect education, skills training and employment opportunities in rapidly changing labour markets.
The Skills2Job Challenge tasked participants with building machine learning systems capable of predicting the five most relevant occupations from just five skills. Using real-world job postings data from UNESCO’s Global Skills Tracker, competitors developed AI-powered recommendation systems designed to improve career guidance, workforce mobility and skills-based hiring.
After taking 1st place at the Skills2Job Challenge, Lucia Yen, a student at Carnegie Mellon University Africa in Rwanda, explains that her winning approach focused on the most overlooked aspect of recommendation systems: retrieval. Rather than concentrating only on ranking models, Lucia designed a two-stage recommendation engine that first retrieved a broad pool of possible occupations, before ranking them using advanced machine learning models.
When a system underperforms, the answer is not always a more complex model. Sometimes the real breakthrough comes from understanding where information is being lost and fixing that bottleneck first. In this challenge, candidate retrieval was that bottleneck.
Breaking an overreliance on formal credentials
The Technical University of Mombasa team – comprised of team members Emmanuel Cherutich, Karuiki Njenga, Bryan Mwaura, Vincent Kututa – known collectively as ‘Team Adventurers’ secured second place. Their solution approached the challenge as a text-classification problem rather than a traditional recommendation system, noting that labour markets in Africa often over-rely on formal credentials while overlooking transferable skills, such as communication, management and sales.
Looking beyond the competition, Team Adventurers see significant potential for AI-driven skills matching.
Across the EAC’s integrated market, where workers increasingly move across borders, a shared skills language could become the common currency of regional employment – fairer, faster and more human than a CV alone.
READ MORE: https://www.unesco.org/en/articles/ai4eac-innovation-challenge-east-african-students-build-ai-could-reshape-african-hiring?hub=701
