How to keep enjoying programming in a world of LLMs

The article offers advice on maintaining enjoyment and productivity in programming despite the rise of large language models (LLMs). It acknowledges ethical concerns about frontier LLMs from big tech companies but is not about them. The author emphasizes that despite fears of losing jobs to unskilled AI programmers, the human role in software development still holds value.

To stay productive and enjoy coding, the author suggests using LLMs for mundane tasks like organizing research, testing, and planning. However, crucial decisions should stay with humans. LLMs are prone to generating buggy code, so it's better to pair them with human coding. The author recommends using LLMs for tasks like generating todos and summaries from conversations, but only after the researcher has familiarized themselves with the domain.

The key takeaway is to leverage LLMs for bookkeeping and routine tasks, while maintaining human involvement in critical aspects of programming. By doing so, programmers can avoid LLM burnout and continue to enjoy and excel in their craft.