Vector Database

In recruiter context

If RAG is the pattern, the vector database is the storage layer that makes it work. Each chunk of source text is converted to an embedding (a vector of floats, typically 384-3072 dimensions) and stored alongside metadata. At query time, the user's query is also embedded, and the database returns the chunks whose vectors are closest by cosine similarity or dot product. Choice of vector DB matters: Pinecone for managed simplicity and scale; Qdrant or Weaviate for self-hosted with rich filtering; pgvector when you already run Postgres and don't want a separate system. Data Engineer and AI Engineer roles increasingly require hands-on vector DB experience.