The increasing importance of recruitment processes, coupled with the growing number of educated yet unemployed individuals, continues to emphasize the vital role of an accurate hiring process. This highlights the significance of the machine learning-based job and CV matching system addressed in this study. The system automates the evaluation of resumes and job postings by using machine learning algorithms such as Natural Language Processing (NLP), Random Forest, LightGBM, and Logistic Regression. Two primary datasets, consisting of resume and job posting data, were prepared through web scraping and synthetic data generation methods. The data underwent preprocessing steps such as text normalization, feature extraction, and labeling to ensure consistency and accuracy. Advanced feature extraction techniques, including TF-IDF, Bag of Words (BOW), and Word2Vec, were employed to analyze the relationship between resumes and job descriptions. Model performance was evaluated using metrics such as precision, recall, and F1-score. LSTM outperformed both LightGBM and Logistic Regression, achieving the highest values in accuracy (0.9526), precision (0.9753), recall (0.9275), and F1-score (0.9508). This indicates that ensemble methods and gradient-boosting techniques are particularly effective for job-matching tasks. The findings reveal the significant potential of machine learning models to enhance recruitment workflows.