The rapid growth of online learning platforms has increased the demand for accurate course recommendation mechanisms and secure digital certification systems. Existing e-learning frameworks typically rely on hybrid filtering, clustering, or ontology-based approaches for personalization and certificate management. However, these systems struggle with dynamic course structures, learner diversity, and weaknesses in centralized certificate verification. To address these challenges, this paper proposes an XGBoost–Integrated Blockchain Certifier Framework that combines a lightweight course recommendation model with a secure blockchain-based digital certification mechanism. The XGBoost model analyzes learner academic history, performance indicators, and skill preferences to generate personalized course recommendations, achieving a correlation value close to 0.99 with low prediction error across multiple training configurations. For certificate security, the framework employs an extended X.509 certificate structure integrated with blockchain-based hashing, ensuring tamper-proof validation and transparent verification. Experimental results demonstrate strong predictive accuracy, stable generalization, and reliable certificate integrity under increasing user load, making the proposed system suitable for modern e-learning environments.