Vector DB e RAG

8 pages in this section.

How Vector Search Gives Agents a Searchable Memory

Explains how embeddings and similarity search let an agent retrieve relevant past information without stuffing every document into the context window.

Python Vector Db Rag Memory Explainer Embeddings

Vector Databases & RAG Basics

A first embedding, in-memory index, and similarity search over a handful of documents, then intermediate steps for metadata filters, persistence patterns, and a minimal retrieval function an agent can call.

Python Vector Db Rag Memory Basics Vector

Choosing a Vector Store: Pinecone, Chroma, and pgvector Compared

Compares managed and self-hosted vector store options - Pinecone, Chroma, and pgvector - by scale, cost, and ops overhead for agent RAG workloads.

Python Vector Db Rag Memory List Ranked Decision

Building a Retrieval Tool an Agent Can Call

A recipe for wrapping vector search as a callable retrieval tool with a clear schema, filters, and structured results the agent can cite.

Python Vector Db Rag Memory Building Retrieval

Chunking Strategies That Affect Retrieval Quality

Python Vector Db Rag Memory Explainer Advanced

Hybrid Search: Combining Vector Similarity with Keyword Filters

Python Vector Db Rag Memory Advanced Hybrid

Re-Ranking Retrieved Results Before Passing Them to an Agent