AI
Embeddings
Numerical fingerprints for chunks of text that let you find semantically similar passages — the search layer underneath RAG.
An embedding turns a chunk of text into a vector — a list of numbers — that captures its meaning. Two passages about 'replacing the throttle cable' end up with similar vectors even if they use different words.
Embeddings are the retrieval half of Retrieval-Augmented Generation. You embed your corpus once, store the vectors in a vector database, and at query time you embed the user's question and find the nearest matching chunks to feed the model.
Quality of embeddings + quality of chunking = quality of your AI answers. This is where 'have a clean markdown corpus' starts to matter a lot.