The rapid growth of digital learning platforms has exposed limitations in traditional assessment systems that primarily rely on score-based evaluation without providing meaningful feedback. This paper proposes an Explainable AI-based Intelligent Learning and Assessment System that integrates performance analytics, semantic retrieval, and explainable feedback to enhance personalized learning. The system evaluates student responses to identify topic-wise weaknesses and utilizes vector embeddings to retrieve contextually relevant study materials. A Retrieval-Augmented Generation (RAG) framework is employed to generate human-readable explanations, improving transparency and conceptual understanding. Additionally, a gamification mechanism is incorporated to enhance learner engagement through experience points and achievement levels. The system is designed using a scalable backend architecture that ensures efficient data processing and secure user interaction. Experimental results indicate improved recommendation relevance, accurate weak-topic detection, and enhanced learning outcomes. The proposed system provides an integrated and explainable solution for modern intelligent learning environments.