Vector Databases

Vector Databases

Vector databases store embeddings and enable fast approximate nearest-neighbour search to find similar documents.

# Chroma (local dev — no API key needed)
import chromadb

client = chromadb.Client()
coll = client.create_collection('docs')

coll.add(
    documents=['Docker containers are isolated environments',
               'Kubernetes orchestrates container workloads'],
    embeddings=[[0.1,0.2,...],[0.15,0.25,...]],
    ids=['doc1','doc2'],
    metadatas=[{'source':'docker.md'},{'source':'k8s.md'}]
)

results = coll.query(
    query_embeddings=[[0.12,0.22,...]],
    n_results=2
)

# Pinecone (production managed)
from pinecone import Pinecone
pc = Pinecone(api_key='your-key')
index = pc.Index('my-index')
index.upsert(vectors=[{
    'id':'doc1',
    'values':embedding,
    'metadata':{'text':'...','source':'docker.md'}
}])
results = index.query(vector=q_embed, top_k=5, include_metadata=True)

# Options: Chroma (dev) | Pinecone | Qdrant | Weaviate | pgvector