This study examined the integration of AI-driven learning analytics and automated feedback mechanisms to address persistent challenges in educational measurement and evaluation in Imo State public tertiary institutions. Conducted at Alvan Ikoku Federal University of Education, Owerri and Imo State University in Imo State, the research employed a convergent parallel mixed-methods design embedded within a quasi-experimental framework. The target population consisted of undergraduate students and academic staff in selected departments. A sample of 400 students and 20 lecturers participated in the six-month intervention. The study first assessed current manual assessment practices, which were characterised by delayed feedback, inconsistent marking, and high workloads. A custom AI-powered platform with locally hosted models was then designed and deployed, featuring automated scoring, real-time personalised feedback, and learning analytics dashboards. Data were collected using pre/post achievement tests, system logs, semi-structured interviews, and a Technology Acceptance Model survey. Results showed significant improvements with the AI system: mean test scores rose from 58.4 to 72.6 (+14.2 points, p<0.001), grading time decreased by 89%, assignment submission rates increased from 67% to 91%, and feedback delivery improved from 21 days to under 5 minutes. Student engagement reached 89%, with high adoption rates, particularly at Alvan Ikoku Federal University of Education, Owerri. The findings demonstrate that AI integration can enhance assessment quality, timeliness, and learning outcomes in resource-constrained settings. The study offers practical recommendations for policy and scalable implementation across Nigerian tertiary institutions.

Keywords: AI-driven learning analytics, automated feedback mechanisms, educational measurement and evaluation, Nigerian tertiary institutions.