This study optimized the polyphenol extraction conditions for Cassia fistula L., commonly known as the Osaka tree. Its fruit contains a complex of valuable natural compounds, which remain underexploited in Vietnam and other countries where Osaka trees grow in abundance. Methods employed included phytochemical screening, determination of total phenolic and moisture content, preliminary factor screening, and optimization using Response Surface Methodology (RSM) with a Box-Behnken design, followed by statistical analysis using t-tests, ANOVA, and Tukey’s HSD post-hoc test. Most previous studies focused on testing and evaluating individual extraction parameters, such as temperature, solid-to-solvent ratio, and time. This study comprehensively integrated these factors into a unified optimization model. Based on the design of preliminary diagnostic experiments, the solid-to-solvent ratios ranged from 1:10 to 1:50 (w/v), the extraction temperature was 30–70°C, and the extraction time ranged from 5 to 30 min. The Box-Behnken model combined with response surface methodology made it possible to optimize multiple factors simultaneously. The total polyphenol content reached its highest value of 10.64 mg GAE/g DW under the following optimal conditions: 50°C; 21.15 min extraction time; 1:31.69 (w/v) solid-tosolvent ratio. The model demonstrated high reliability (R2 = 0.94), confirming its industrial potential. The results obtained open up new prospects for sustainable exploitation of natural polyphenols from C. fistula in functional foods and pharmaceuticals.