Generative AI tools like GitHub Copilot and OpenAI’s ChatGPT are reshaping software development by automating coding tasks and influencing design approaches. As these technologies become more prevalent, computer science education must evolve to prepare students for this changing landscape.

Our Object-Oriented Design course, offered as part of a master’s bridge program, integrates AI tools with two key objectives. The first is to engage students with AI in a structured way, preventing superficial understanding and potential misuse that may arise from unguided exploration. By introducing these tools within the curriculum, students learn how to use them effectively while being mindful of their limitations. The second objective is to align the course with industry practices, where AI-assisted development has become standard. Familiarity with these tools ensures that students are equipped to navigate AI-enhanced development environments.

The course is structured into two distinct phases. The foundational phase focuses on Java programming, object-oriented principles, and unit testing, intentionally omitting AI tools to help students develop critical thinking skills and a solid grasp of programming fundamentals. In the design phase, AI tools are introduced to automate repetitive coding tasks, allowing students to concentrate on higher-level design challenges and reinforcing the emphasis on software design over coding mechanics.

To ensure a smooth transition to AI-assisted learning, we incorporate guided instruction on AI prompting, evaluation, and recognition of tool limitations. Critical thinking is emphasized by encouraging students to verify and validate AI-generated code rather than relying on it uncritically. Ethical considerations are also addressed, with discussions on responsible AI usage, ownership, and plagiarism forming a key part of the curriculum.

Assessment extends beyond traditional code evaluation by incorporating structured code walks. For homework assignments, students record short video presentations explaining their code, a process that reinforces comprehension, improves communication skills, and fosters confidence. Final projects require live, synchronous code walks, where students present their Unified Modeling Language (UML) diagrams, discuss their design choices, demonstrate unit tests, and reflect on challenges. These interactive sessions provide immediate feedback from instructors and teaching assistants while encouraging students to articulate their reasoning and consider improvements.

The integration of generative AI tools and structured code walks has enhanced student learning, allowing them to tackle more complex software projects. Looking ahead, efforts will focus on developing scalable assessment models, such as expanding the use of recorded sessions and training teaching assistants to maintain consistency in code walk evaluations. Additionally, research will be conducted to assess long-term student outcomes, ensuring that the course continues to evolve in response to educational and industry needs.

By aligning education with industry trends, this approach ensures that students develop both foundational programming expertise and proficiency in AI-assisted design, preparing them for the future of software development.

Keywords: Technology, education, artificial intelligence, generative ai, coding, design.