What are the current and planned memory for AI agents in LlamaInmdex?

Does LlamaIndex Workflow work with Redpanda? Or it needs to be confluent?

What is the interface to LlamaIndex workflow? Does it have an API for client to access? Is there a client sdk for it?

What are the tradeoffs to doing rankings/reranking inside vs outside the VectorDB? Some VectorDBs have extensive ranking support.

Is sentence splitter still optimal for embedding models like bge-m3 that can vectorize a whole article or paragraph?

What is the default behaviour of load_data() on SimpleDirectoryReader Surprised to see it return 9 items from a directory with just one readme.md file! required_exts is just .md is it auto parsing into markdown nodes?

Hybrid search is now RAG best practice so with Qdrant as vector store what is difference between https://docs.llamaindex.ai/en/stable/examples/vector_stores/qdrant_hybrid/ and using Fastembed?

2) when using a lllamaindex doc loader I want to see what files loaded or did not load. What log settings etc is best for this. For example I used an .md loader in a directory of sub directories and it loaded 80% I need to know what skipped and why

I may use the cloud later but now I just want to use best pdf parser library

What is current state of Llama-Index integration with Vespa? I just about ready to upload my vector embeddings and integrate with RAG pipeline

And what is the order in which nodes are returned from the parser for a single document?