At Ford, machine learning touches every corner of the business—this includes banking, product development, manufacturing, customer experience, quality, and more. Since Ford's machine learning needs are wide-reaching, Ford deploys Machine Learning pipelines in a wide variety of applications—from computer vision and anomaly detection to natural language processing. In this talk, Austin Denha, Machine Learning Engineer and Product Owner for Machine Learning Platforms at Ford, will demonstrate how he and his team built generalized pipelines to democratize the deployment of models across the company. He will walk through his views on what is necessary to succeed in this kind of project, from the importance of clean data to the requirements of model training and deployment tools. Austin will also explain his philosophy and approach the teams at Ford have adopted to get machine learning pipelines into production. Importantly, he will discuss their production first mindset and how they enabled data scientists to manage their deployments and bring the time-to-deploy from months to days.