This study investigates the prediction of interview decisions by integrating facial features with Five Factor Model (FFM) personality traits in a cross-modal framework. We compare three approaches—the facial-feature model, the personality-based regression model, and the proposed integrated model that concatenates deep facial representations with FFM personality traits scores—using a public dataset of short interview clips. Experimental results demonstrate that the proposed integrated model outperforms the facial-feature model, achieving lower error rates and higher correlation and concordance metrics when predicting interview outcomes. These findings indicate that combining internal psychological traits with external appearance cues enhances the robustness and accuracy of automated interview assessment tools, which offers a more comprehensive basis for making fairer and more efficient hiring decisions.