As developers we’ve all heard a lot about Machine Learning, Big Data and I'm certain that most if not all of us would love to get involved in this exciting field. However the pressures of juggling the demands of a full-time career in software can make getting started in such a broad field feel a little daunting. I’d like to change that! What I’d like to cover in this talk are the fundamental concepts of Linear Regression. Next we’ll look at Gradient Descent, a simple, yet powerful technique we can use to find the ‘best fitting’ parameter values for our model. Next, a we’ll take a quick look at Logistic Regression and discuss the central tenant of any Neural Network, the Perceptron. With this understanding, we can jump into building and initializing a small Neural Network. The last piece of the puzzle is how we train a Neural Network. We’ll discuss Back Propagation, an algorithm that retrospectively updates the weightings between nodes our Neural Network. With that knowledge in hand, we’ll have everything required to throw some unseen data at our Neural Network and watch the magic unfold. Last but not least, is a look at how Neural Networks can be used in practice. Whilst the solve many real life problems, from enabling an autonomous vehicle to navigate a busy highway to detecting fraudulent banking patterns, for the purposes of this talk I’ll be demonstrating how a neural network can be used to solve the age old problem of handwriting recognition. By attending this talk attendees will walk away with a clear understanding of the deceptively simple techniques used to build the majority of the learning algorithms used in industry (Linear Regression, Gradient Descent). My goal is to motivate attendees to seek ongoing education in this space, and at the conclusion of the talk references will be provided to material that I’ve found invaluable in my personal journey, including a brief review of Coursera’s Machine Learning MOOC.