It's a bit challenging for me to post here, so please bear with my English.
I'm a junior software engineer in Tokyo, and I'm learning software development on my own.
Recently, I started building software that uses LLMs and AI agents. I understand that in conventional software development we can follow the red-green-refactor cycle, but I'm wondering how people evaluate software built with LLMs or AI agents.
How do you test or evaluate these kinds of applications? Do you have any advice or best practices?
Top comments (0)