Reclaiming the White Spaces: An AI-Powered Micro-Learning Framework for Practical Skill Acquisition Within University Timetable Gaps in African Higher Education

African higher education faces a persistent skills mismatch: graduates emerge academically certified but practically underprepared for the technology workforce. This paper presents Octal Foundry, an AI-powered micro-learning platform designed to help African university students acquire practical technology skills within the gaps between their scheduled lectures. Over four weeks of research, we conducted a systematic literature review of 17 academic sources, a competitive analysis of six platforms, technical evaluation of Retrieval-Augmented Generation (RAG) pipelines and API strategies, an analysis of eight university curricula across four Kenyan universities, a privacy compliance review against the Kenya Data Protection Act 2019, a small model feasibility study for offline AI capability, and a survey of 50 university students. Key findings include: micro-learning in 15-45 minute sessions improves knowledge retention by 20-30 percent; 82 percent of surveyed students are likely or very likely to use a platform that ingests their timetable and generates personalized learning roadmaps; the biggest frustrations for self-directed learners are the absence of a clear roadmap (32 percent) and lack of motivation (28 percent); and Kenyan STEM students have an average of 17 hours of white space per week approximately 60 percent of their total class time. The proposed architecture operates at near-zero infrastructure cost for a pilot of up to 500 students.