Articles discussion: LLMs and the future of software engineering

Goal: Read all three of the below articles. How will the information in the articles affect the way you approach the CS169L project, and your other software project(s) generally? Write a short summary of your thoughts as part of this assignment (individually, not in teams), and we'll have an informed discussion about it on Monday.

Article: What do LLMs mean for the future of software? Links to an external site.

Bertrand Meyer is a distinguished software engineer known for his work on formal methods, including software contracts. Here's his take on the difference between code generation and software engineering, and implications for the integration of AI into SW Engineering activities.

Article:  What we learned copying all the best code assistants Links to an external site.

Which AI tools are best at which kinds of programming-related activities? What are some concrete examples of using them to generate pieces of code that integrate into larger projects?

Article: Using LLMs and Cursor to become a finisher Links to an external site.

What if you use an LLM to help you develop a spec for the code you need written, then use a different LLM session to turn that spec into code?

Substack post: Prompt Driven Development Links to an external site.

The author identifies PDD, in which the "developer" doesn't write any code themselves but does everything else, as particularly suitable for people who "likely have an architectural understanding of how software works, but have not retained fluency over programming syntax."