-
Book Overview & Buying
-
Table Of Contents
Practical C# Projects with .NET
By :
Often, when you're building an AI application, you'll encounter scenarios where an LLM on its own is not enough to get the correct response from your agent.
LLMs are trained on very large bodies of text, and they possess no reasoning capabilities of their own. This means that it's very possible for users to ask your agent questions and get blatantly incorrect or incomplete information.
Prompts can help with this, but it's often impossible or at least impractical to put every conceivable piece of information into a system prompt or a few-shot inferencing example.
To combat this, AI engineers can bolt on additional pieces of logic into their applications that provide tools in the form of functions that are then made available to the LLM. LLMs that support tool calling can request that these functions be called. The functions can then be called and their results returned to the...