Nolan and Temple Lang's "Computing in the Statistics Curricula" (2010) advocated for a shift in statistical education to broadly include computing. In the time since, individuals with training in both computing and statistics have become increasingly employable in the burgeoning data science field. In response, universities have developed new courses and programs to meet the growing demand for data science education. To address this demand, we created Data Science in Practice, a large-enrollment undergraduate course. Here, we present our goals for teaching this course, including: (1) conceptualizing data science as creative problem solving, with a focus on project-based learning, (2) prioritizing practical application, teaching and using standardized tools and best practices, and (3) scaling education through coursework that enables hands-on and classroom learning in a large-enrollment course. Throughout this course we also emphasize social context and data ethics to best prepare students for the interdisciplinary and impactful nature of their work. We highlight creative problem solving and strategies for teaching automation-resilient skills, while providing students the opportunity to create a unique data science project that demonstrates their technical and creative capacities.