Support Vector Machines are a supervised learning method developed by Vladimir Vapnik and his colleagues, with the modern formulation published in 1995. They work by finding the boundary that best separates classes of data, maximizing the margin between them, a principle that delivered the best accuracy on high-dimensional problems like handwriting recognition, face detection, and text classification. Through the 1990s and 2000s, they became the dominant paradigm of machine learning, the state of the art before deep learning arrived, built on clever mathematics and engineered features rather than networks learning their own representations. AlexNet's win in 2012 ended that reign, which is why telling the story of the modern era requires knowing what came before it.