From the simplest models to complex deep neural networks, modeling turbulence with machine learning techniques still offers multiple challenges. The advent of new powerful deep learning methods have fostered their applications in a wide range of research areas, including more recently in fluid mechanics. From the simplest models to complex deep neural networks, modeling turbulence with machine learning techniques still offers multiple challenges. This seminar will cover some of the ideas of deep learning applied to computational fluid dynamics. Furthermore, the capabilities of deep learning methods to perform various predictions in turbulent flows would be explored including the use of convolutional neural networks for predicting the turbulent flow past an obstacle. The potentials of robust deep learning strategies using patch-based learning would be presented along with its application to learning a turbulence model.