AI as a Pair Programmer

I do think AI/LLMs have value, but not as much as the “AI enthusiast crowd” thinks. I like specific concrete examples of what AI can do. I think that treating AI as a pair programmer is one case where it provides value.

I don’t have much experience with human pair programming, but I have seen that having multiple people design and review code as it is written can be fantastically productive — and sometimes not. If you are a single developer, you can use AI as your pair programmer by asking it to explain and review code, suggest alternatives, and look for ways to simplify. In this setup, you are still the developer: the AI may write code, but you still need to understand it. Used this way, AI challenges your ideas, helps you learn, and leads to better code.

Specifically, I have used the paid versions of Copilot CLI and Gemini CLI with frontier models and I have used Pi with open and local models. They work similarly for my pair-programming workflow. Smaller models may still be useful: they may have less capability and older training data, but they can still provide helpful suggestions. Even the smallest local model I have tried, Gemma 4 e2b, has (slowly) produced good suggestions on my non-GPU machine.

In short: I think AI pair programming improves code, and it creates a learning environment for the developer with good access to information and faster, better feedback. I think you should have AI work together with you when you write code you care about¹.