SE Radio 582: Leo Porter and Daniel Zingaro on Learning to Program with LLMs
Software Engineering Radio - the podcast for professional software developers
The Fallibility of the LLM
Students tend to trust their code without testing it because they wrote it, but this can lead to bugs. By using a large language model to generate code, students are more likely to engage in testing practices. Test driven development can become the default approach, with students starting by writing tests and then having the model generate the code. This encourages students to think about the desired behavior of their functions and the expected inputs and outputs.
00:00
Transcript
Play full episode
Remember Everything You Learn from Podcasts
Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.
