This project aims to develop effective strategies for human-AI collaboration in responsible software engineering, advancing a human-centered hybrid intelligence approach that integrates AI capabilities with human expertise and values. Motivated by the rapid adoption of AI in software development—and the growing risks of misalignment with human and societal values—the research investigates how to preserve human agency, inclusivity, and accountability in AI-assisted systems. The project will have both scientific and societal impact by producing empirically grounded best practices, theories, and educational resources that support ethical and effective AI integration in software develpoment. Methodologically, the research adopts an iterative, mixed-methods approach in which best practices and insights are empirically developed and evaluated through real-world, community-engaged software projects within the INSPIRE program—a university-led initiative that brings together multidisciplinary student teams to co-design AI-enabled solutions for societal challenges in partnership with community organizations. This setting enables close collaboration with diverse stakeholders and validation in authentic development contexts.