I’m working on building multiple agents (system and user prompt-based) that need to dynamically select the appropriate RAG catalog based on user inputs and specific parameters. I’m exploring the best approach for routing this traffic—whether using a decision tree, LLM with system prompts, or another method. Given the complexity and nuance of human language, it’s challenging to fully understand the context during interactions, which makes the task difficult. Any advice or suggestions would be greatly appreciated.
Thanks in advance!