Dengue virus represents a persistent global health burden, and the absence of clinically approved antiviral therapies highlights the urgent need for novel treatment strategies. The viral NS2B/NS3 serine protease, essential for polyprotein processing and viral replication, constitutes an attractive target for structure-based drug discovery. The presented study is an attempt to repurpose FDA-approved drugs targeting the allosteric site of DENV2 (PDB: 6MO0) and JEV (PDB: 4R8T) NS2B/NS3 proteases. FDA-approved drugs library from ZINC database was subjected to structure based high-throughput virtual screening using AutoDock-GPU in Google Colab. The drugs commonly appearing at top 100 HITs in both the screens were extracted and further subjected to molecular dynamics (MD) simulation to identify the stability of the complexes and their corresponding estimated free energy of binding using AMBER. Identified drugs are currently under procurement and will be evaluated for antiviral activity against DENV2 and JEV in a cell based antiviral assay. Parallelly, an attempt has been initiated to develop a QSAR model using the DENV2 allosteric inhibitors reported till date using mcQSAR tool. The model will be used to predict the activity of identified HITs. This integrated computational approach is expected to accelerate the identification of promising antiviral candidates for subsequent experimental validation against flaviviral infections.