Esophageal squamous cell carcinoma (ESCC) is known as low early diagnosis rate and poor prognosis survival. Effective prognostic markers are urgently required to predict outcome for ESCC treatment. Accumulating evidence supports a strong relationship among cancer cell growth and amino acid metabolism. This work focused on the comprehensive evaluation of the association between amino acid metabolism-related gene (AAMRGs) expression profiles and prognosis in ESCC patients based on The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO). 2 AAMRGs (BCAT1 and MMACHC) were selected to develop a novel AAMRGs -related gene predictor for 1-year and 2-years prognostic risk in ESCC patients. Both BCAT1 and MMACHC expression were verified by RT-qPCR. Prognosis nomogram by incorporating clinical factors and BCAT1 and MMACHC gene expression was constructed and the calibration plots illustrated good prognostic performance. In summary, the AAMRGs signature presented in our study is efficient and holds potential for clinical use in forecasting the prognosis of ESCC.