Does it really know about the whole codebase? How big are we talking about? Last I tried, some time ago, llm didn't work very well "in the large". But I have never tried Claude Code.
No it doesn’t, it tries to be smart about how it loads context that it thinks it needs to perform a task. In reality, in a large codebase I’ve found manual supplying the exact context to Aider with Anthropic’s models to work better.
Same. It's usually very good at generating greenfield projects from scratch, but once you get going you have to manually provide all context to get good results.
I have tried claude code on a few different size code bases and so far very impressed. The code it generates is usually in the same fashion and tone as the rest of the code base.
I guess I should have said that it has access to all your code. But I think it tries to be smart about which files it reads depending on the task at hand.