Purpose of the study: This study aims to analyze the validity and readability of a deep learning–based electronic module (e-module) designed to improve students’ critical thinking skills on global warming material. Methodology: The research employed a Research and Development (R&D) approach using the ADDIE model, consisting of analysis, design, development, implementation, and evaluation stages. The developed e-module integrates deep learning principles—mindful, meaningful, and joyful learning—into a Google Sites platform and is enriched with multimedia elements, including videos, animations, interactive worksheets, Padlet, and a carbon footprint calculator. The validity of the e-module was evaluated by six experts, including media experts, subject matter experts, and linguists, using Aiken’s V formula. Main Findings: The results showed that all validation aspects achieved Aiken’s V values ranging from 0.83 to 1.00, exceeding the minimum validity threshold (V > 0.78), indicating that the e-module is highly valid. Readability testing involved ten junior high school students and two science teachers and was analyzed using percentage-based criteria. The readability scores ranged from 88% to 100%, categorized as excellent across content, format, presentation, and language aspects. These findings indicate that the deep learning–based e-module is not only valid but also highly readable and user-friendly. Novelty/Originality of this study: The developed e-module has strong potential to be implemented as an innovative digital teaching material to support science learning and foster students’ critical thinking skills, particularly in addressing contextual and global environmental issues such as global warming.