With the frequent changes in technology, the process of hiring software engineers has been changed. The recruiters are now more concerned for the skills (in line with market trends), rather than qualification credentials. Software engineers find it difficult to stay updated with all these quick technological developments. Thus, this study aims to recommend a data-driven ensemble learning model, recommending the prevailing skill requirements to job seekers with the help of examining job descriptions and job responsibilities. This ensemble learning model will incorporate the best word embed- ding techniques (TF-IDF, Glove Wikipedia and Twitter), feature selection techniques (Rank-Sum, Cross-correlation, Rank-Sum with Cross-Correlation, Chi-Square and Principal Component Analysis (PCA), and classification techniques (BAGND, BAGL, BAGK, RF, EXTR, AdaB, GRaB, XGBC, LGBMC, MLPL, MLPS, and MLPA). The findings of this study will offer significant recommendations for career preparation, collaborations with relevant industries, and accumulating the right skills with the right role in the software engineering industry. The study will also benefit academia for the upgradation of professional curricula.