Renal transplantation ensures advantages for patients with end-stage kidney disease. However, in some cases early complications can lead to allograft dysfunction and consequently graft loss. Creatinine is a poor biomarker for kidney injury due principally to its inability to help diagnose early acute renal failure and complete inability to help differentiate among its various causes. Markers of kidney damage: kidney injury molecule (KIM1), neutrophil gelatinase-associated lipocalin (NGAL) and beta2microglobulin (B2MG) may ease early diagnosis of graft dysfunction. The aim of this study was to assess serum concentrations of these biomarkers in relation to classical markers of kidney function (creatinine, and cystatin C) and to analyze their usefulness as predictors of kidney damage with the use of artificial intelligence tools. We included 19 patients who had their first kidney transplantation (5 females, 14 males), without prior immunization, having complete HLA typing and a negative cross-match test before transplantation. We determined serum creatinine and Cystatin C and several biomarkers (KIM1, NGAL and B2MG) at 24h post-transplantation. The data was used to build a Random Forest Classifier (RFC) model of renal injury prediction. The RFC model established based on 2 and 3 input variables, KIM1 and Cystatin C, respectively KIM1, NGAL and B2MG, were able to effectively assess the rate of patients with graft dysfunction. With the use of the RFC model, serum KIM1, NGAL and B2MG may serve as markers of incipient renal dysfunction in patients after kidney transplantation.