Hepatocellular carcinoma (HCC) is the most common malignant form of primary liver cancer and one of the leading causes of liver-related deaths. Risk estimation of HCC depends on a matrix of several risk factors, from clinicopathological to gene expression features. From an etiological perspective, the survival prospects of HCC vary from patient to patient. Therefore, a comprehensive study that captures complex relationships among risk factors is one of the ways to quantify prognosis and/or diagnosis, thereby harnessing personalized medicine. We integrated survival analysis with bioinformatics to identify diagnostic and prognostic bio-markers in HCC. Using RNA-seq data from TCGA-LIHC, we screened differentially expressed genes (DEGs) and applied Cox regression, machine learning (LASSO-penalized modeling), and pathway enrichment analysis to prioritize genes linked to overall survival (OS) and tumor progression. Among 3,623 DEGs (685 downregulated, 2,938 upregulated), we identified three core prognostic genes (AP1M1, CASC9, and CA9) significantly associated with OS (logrank p<0.001, hazard ratio > 2.0). Diagnostic biomarkers (CLEC4M, SLC5A1, SLC22A1, SFRP5, CNDP1, TERT, and ZIC2) were enriched in metabolic signaling pathways. Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis revealed five significantly altered pathways, with the strongest enrichment in ‘hsa04976: Bile Secretion’, underscoring disrupted metabolic homeostasis as a hallmark of HCC aggressiveness. Our study reveals novel biomarkers with both diagnostic and prognostic potential, offering a framework for targeted HCC management. These findings highlight the potential of integrative omics-survival modeling to advance precision oncology.