While much attention is being paid to incorporating GenAI into computing education, most of it has focused on using it for programming-centric tasks, such as code generation, giving feedback on code, or providing synthetic programming partners, all of which are important basic skills and are amenable to automated grading. However, in upper-division software engineering and project courses, interpersonal skills (such as team meetings or customer interviews) are equally important but instructor-intensive to teach. GenAI presents the possibility of scaffolding the teaching of some of these practices. This shift puts the emphasis on tools specifically designed for exercises that encourage students to develop the ability to investigate a topic by iteratively asking questions to find a solution. The goal is to create scenarios where students must train to interact with humans in real-world situations, simulating these interactions in a controlled, guided environment. These simulations could help students practice and refine soft skills, such as conducting interviews or working in teams, by mimicking the types of exchanges and problem-solving they would encounter in professional environments. This approach allows learners to engage in realistic communication exercises, improving their ability to handle complex, interpersonal tasks through repeated practice with AI-guided feedback. As an example, we envision examples that include requirements elicitation with customers, development team meetings, discussion with potential investors, to name just a few.