This article examines modern approaches to enhancing borrower creditworthiness assessment systems, a key tool for improving the quality of a commercial bank's loan portfolio. The study is driven by shifts in the structure of bank lending, the evolution of digital financial technologies, the expansion of remote lending, and the need for more precise credit risk identification. The research aims to develop a comprehensive approach to assessing borrower creditworthiness by combining traditional financial analysis, credit scoring, cash flow analysis, alternative data, artificial intelligence methods, and continuous monitoring of the client's financial status. The study employs methods of systems, comparative, and structural-functional analysis, alongside a synthesis of scientific approaches and an examination of international banking practices. It identifies the primary shortcomings of traditional creditworthiness assessment systems and outlines avenues for their improvement. A multi-level borrower assessment model is proposed, incorporating financial, non-financial, behavioral, and digital indicators. The study demonstrates that improving creditworthiness assessment not only enhances the quality of individual lending decisions but also improves overall loan portfolio management by reducing default probability, enabling the early identification of troubled borrowers, and optimizing the credit risk structure.