Vectorisation is the technique of transforming text into numerical vectors that machine learning algorithms can interpret. Embeddings, on the other hand, are a more advanced type of vectorisation that capture the semantic relationships and contextual nuances between words. These methods allow language models in AI to understand the subtleties of human language and generate relevant, human-like responses.

Hermes utilises the cutting-edge of these technologies to produce indices of data to use as a "source of truth" for any large language model, such as ChatGPT. This can produce incredible results when used in conjunction with a live chat interface, as while the language model understands the customer, the index understands the business.