Let’s develop more ML systems that solve real-world problems! But first, we invite you to explore how much they can differ from solutions built for a competition or during research. When ML transferred from a purely academic discipline to a technology that is actively being implemented in business, specialists started to experience new unfamiliar challenges related to data processing, algorithm selection, communications with the customer, expectations management, and many more. This talk will make you a little more prepared for these issues, whether you are a developer or an entrepreneur. We will look at two projects implemented in production: carotid artery examination and vehicle behavior analysis. Going through some significant challenges faced during the development, we will analyze how they were tackled, and what are the general options to address these types of problems. We expect these examples to help you deal with problems you might encounter in your future real projects.