This is a great question, and it's we can always discuss in slack, doesn't have to be during the live event!
There is overlap between them, most people (and the book) mention Llama index is really great when it comes to connecting LLMs with various data sources.
LangChain now seems to kind of have abstractions for almost everything and integrates with their other products, like LangSmith and LangServe. But I also know quite a few people who switched away from LangChain after the prototyping phase. Sometimes the abstractions feel like a lot 😅