SE Radio 582: Leo Porter and Daniel Zingaro on Using LLMs in the Classroom

Skill Development

Large language models (LLMs) are transforming how programming skills are taught, emphasizing critical skills like debugging and problem decomposition. Daniel Zingaro highlights the importance of students verifying code correctness through strong debugging skills, as LLMs can generate code with errors 1. Leo Porter adds that LLMs encourage students to engage in problem decomposition, a skill often overlooked due to the focus on syntax in traditional courses 2.

LLMs are now giving us the ability to teach essential skills that aren't core topics in many introductory CS classes.

--- Leo Porter

These models necessitate a shift from syntax-focused learning to understanding code structures and logic, preparing students for real-world software engineering challenges.

   

Student Engagement

LLMs are also enhancing student engagement by providing meaningful learning experiences and reducing the temptation to cheat. Daniel Zingaro argues that authentic tasks, like creating a landing page, motivate students more than decontextualized assignments 3. Leo Porter believes that LLMs can make programming accessible to a broader audience, including those who take only one programming class 4.

If you teach them the skills of interacting with these LLMs, they'll be able to do these tasks later.

--- Leo Porter

By focusing on real-world applications, LLMs help students see the value in their work, fostering a deeper connection to the material and encouraging long-term retention of programming skills.