We surveyed developers on their use of LLMs for development tasks, and the ways AI helps or hinders their work. Here's what they told us.
take a critical eye to the uses of AI for development tasks.
Our respondents cover:
- debugging,
- writing unit tests,
- research and planning,
- refactoring code,
and much more. We show how the effectiveness of LLMs varies depending on not only the nature of the task, but the approach of the user.