This report describes the design and implementation of a Proof of Concept (PoC) for an Automated Machine Learning (AutoML) tool aimed at supporting model validation within Société Générale’s Model Risk Management (RISQ/MRM) division. The tool automates the exploration of alternative machine learning pipelines—covering preprocessing, model selection, and hyperparameter optimization—thereby accelerating the review process and improving analytical depth. Built in Python with a modular and extensible architecture, it integrates seamlessly into existing workflows and supports binary and multiclass classification, plus regression. Key features include configurable experiments, interactive dashboards for interpretability, and reproducible pipelines. The PoC demonstrates that AutoML can significantly reduce validation time, enhance transparency, and strengthen governance in regulated environments, while offering a scalable foundation for future extensions such as distributed computing and advanced model types..