Thoughts on Advanced Chunking Strategies for RAG

I’ve been thinking a lot recently about the "chunking problem" in Retrieval-Augmented Generation. If you've played around with the llm CLI tool or built anything with vector embeddings, you've probabl