Evaluating RAG and Semantic Search Systems

As adoption of LLMs and retrieval-augmented generation (RAG) becomes more prevalent in enterprise, the demand for robust evaluation arises, ensuring these new tools meet the high regulatory standards of various industries. Gain an in-depth understanding of the unique requirements as well as practical tools to robustly evaluate such systems addressing the complex nature of the underlying data. We'll introduce evaluation techniques addressing domain-specific language and knowledge, specific evaluation metrics, and independent assessment of the retrieval and generation steps, all while taking temporal information into account.
Prerequisite(s):

Python.
Basic knowledge of LLMs and RAG.