How does an LLM decide which tool to call? If you give a model 50 tools, what happens to its accuracy and context window?

In an API inference call, what is the functional difference between the System, User, and Assistant roles?

How do you prevent a user from using your "Search Tool" to look up internal sensitive data they shouldn't have access to?

How would you set up an experiment to prove that a new prompt version is actually 10% better than the old one? What metrics would you track?