With the expansion of the knowledge-based economy, predicting the job performance of knowledge workers at the early stages of recruitment has become a critical challenge in human resource management. This study aims to identify the key dimensions and components involved in predicting the performance of knowledge worker applicants and to propose an integrative framework to improve hiring decisions in innovative organizations. Using a qualitative–integrative approach, a meta-synthesis of 31 international studies published between 2015 and 2025 was conducted, and performance-related components were extracted through thematic analysis. The resulting framework was then validated using a two-round Delphi method with 10 experts in human resource management and data analytics. The results of the Friedman test ( χ²(29)=145.2, p<0.001) and Kendall’s coefficient of concordance (W=0.75) indicate a strong level of expert consensus on the identified components. The findings suggest that knowledge worker performance is a multidimensional phenomenon that is best predicted through the integration of behavioral and cognitive data, individual characteristics, and human judgment with data-driven analytics. The study offers an evidence-based framework to support intelligent recruitment systems in knowledge-based organizations.