The thermal management of lithium-ion batteries remains a critical challenge for electric vehicle safety and performance. Coupled multi-physics modeling is required to capture complex electrochemical-thermal interactions. This research presents a high-fidelity numerical framework integrating micro-scale electrochemical behavior with macro-scale thermal analysis for Li-ion batteries in liquid immersion cooling configurations. We develop a multiscale modeling approach coupling two physics domains. At the micro-scale, the enhanced Doyle-Fuller-Newman (DFN) model [1] captures ion intercalation, charge transfer kinetics, and temperature-dependent transport properties. The DFN model employs a layer pattern reproducing the jelly roll structure of cylindrical cells. This avoids full battery geometry meshing, significantly reducing computational costs while maintaining physical accuracy. At the macro-scale, three-dimensional conjugate heat transfer (CHT) simulations resolve thermal gradients and heat between batteries and the immersion cooling fluid. The bidirectional coupling framework exchanges temperature fields and volumetric heat generation between both scales. The DFN model is exploited by using the PyBaMM library in Python [2]. CHT simulations are carried out by means of STAR-CCM+ software. Coupling operates through Java scripting, enabling real-time data exchange at each time step. Temperature fields from CFD update electrochemical properties, including ionic conductivity and reaction kinetics. Heat generation feeds back through ohmic, reversible, and irreversible mechanisms [3]. This framework enables critical capabilities for battery pack configurations. It predicts thermal evolution with enhanced accuracy by resolving two high-fidelity models and electrochemical-thermal interactions within cells and small pack patterns. It supports parametric studies of industrial interest: electrode thickness effects on thermal response during discharge, separator material influence under fast charging, and chemistry-dependent behavior across cooling strategies.