A code-first, math-light path into machine learning for hands-on learners. Instead of proofs and derivations, this book teaches ML as a craft, through intuition, real datasets, and fully reproducible Python. Every example runs start to finish. Each begins with a single pip command and is heavily commented, so you can follow every step from loading data to interpreting results. You'll work in scikit-learn, PyTorch, and TensorFlow across five parts: foundations, supervised learning, unsupervised learning, advanced methods like CNNs and interpretability, and integrated capstone projects. Case studies use openly licensed data: smartphone activity recognition, credit card fraud, spam detection, traffic sign classification, algorithmic fairness in recidivism, and an end-to-end NVIDIA stock-forecasting pipeline. Ethics notes run throughout. Adopted as required reading for CSC 373 (Machine Learning) at the University of Advancing Technology. Written for undergraduates with basic Python, career-switchers moving into applied ML, and instructors needing reproducible, lab-ready material.