ARCHIVES

The global job market is undergoing rapid transformation, rendering traditional, static career counselling methods—often based on psychometric tests—insufficient for navigating modern technological disruptions. This paper presents Next Step, a technically sophisticated hybrid AI framework that integrates semantic similarity, demand-weighted skill gap quantification, and longitudinal labour market analytics into a dynamic decision support system. Unlike traditional "black-box" recommenders, Next Step utilizes a hybrid ranking function optimized via Bayesian Tuning to balance immediate skill alignment with long-term economic viability. Our methodology employs Term Frequency-Inverse Document Frequency (TF-IDF) vectorization and a novel Skill Gap Index (SGI) to provide interpretable, actionable guidance for upskilling. Empirical validation through simulation demonstrates that the hybrid model achieves a 20% improvement in Precision@5 and a 23% gain in Mean Reciprocal Rank (MRR) compared to similarity-only baselines. By incorporating Explainable AI (XAI) principles and addressing algorithmic fairness, this research transitions from a conceptual prototype to a validated empirical study, offering a robust framework for personalized, demand-driven career guidance in the age of automation.