Course Wiki

Vector Search with PGVector — LLM Zoomcamp Module 2

--- video_url: 'https://www.youtube.com/watch?v=0P54MFyz-mc&list=PL3MmuxUbc_hLZFNgSad56pDBKK8KO0XIv' --- # Vector Search with PGVector

llmz-module-02

LLM Zoomcamp › Module 2: Vector Search › Vector Search with PGVector

Notes

video_url: “https://www.youtube.com/watch?v=0P54MFyz-mc&list=PL3MmuxUbc_hLZFNgSad56pDBKK8KO0XIv”

Vector Search with PGVector

Many real databases can do vector search. Elasticsearch has it, and there are dedicated stores like Qdrant and Chroma. We’ll go with Postgres. Most of us already run it at work, and the data engineering course uses it too. The concept is the same as with sqlitesearch, only the database under the hood changes.

pgvector is the PostgreSQL extension that makes this work. Install it and Postgres can do vector similarity search. On top of that you get the usual production features, like concurrent access, transactions, and large datasets.

We’ll run Postgres with pgvector in Docker.

Starting Postgres with pgvector

Pull the image and start a container:

docker run -it \
    --name pgvector \
    -e POSTGRES_USER=user \
    -e POSTGRES_PASSWORD=pswd \
    -e POSTGRES_DB=faq \
    -v pgvector_data:/var/lib/postgresql/data \
    -p 5432:5432 \
    pgvector/pgvector:pg17

This image has the pgvector extension pre-installed. The -v flag creates a named volume so data persists across container restarts.

Installing the Python client

Install the driver:

uv add psycopg[binary]

Note: if using Zshell use uv add 'psycopg[binary]'

We’ll use psycopg (v3) to connect and run queries. Note: this is different from psycopg2 - psycopg v3 supports conn.execute() directly without creating a cursor.

Preparing the data

We need the FAQ documents and their embeddings.

Here’s what we did in previous units as one script:

from tqdm.auto import tqdm
from ingest import load_faq_data
from sentence_transformers import SentenceTransformer
model = SentenceTransformer("all-MiniLM-L6-v2")
documents = load_faq_data()
texts = [doc["question"] + " " + doc["answer"] for doc in documents]
batch_size = 50
vectors = []
for i in tqdm(range(0, len(texts), batch_size)):
    batch = texts[i:i + batch_size]
    batch_vectors = model.encode(batch)
    vectors.extend(batch_vectors)

Now we connect to Postgres:

import psycopg
conn = psycopg.connect(
    "postgresql://user:pswd@localhost:5432/faq"
)
conn.execute("CREATE EXTENSION IF NOT EXISTS vector")

The second line activates pgvector. The Docker image we started isn’t plain Postgres, it ships the extension inside, and this turns it on. It adds the vector column type and the similarity search operators.

Creating a table

Create a table for storing documents with their embeddings:

conn.execute("""
    DROP TABLE IF EXISTS documents
""")
conn.execute("""
    CREATE TABLE documents (
        id SERIAL PRIMARY KEY,
        course TEXT,
        section TEXT,
        question TEXT,
        answer TEXT,
        embedding vector(384)
    )
""")

The vector(384) column stores our 384-dimensional embeddings from all-MiniLM-L6-v2.

Inserting documents with embeddings

Let’s insert the documents and their vectors into PGVector:

def vec_to_str(vector):
    return "[" + ",".join(str(x) for x in vector) + "]"
for doc, vec in tqdm(zip(documents, vectors), total=len(documents)):
    conn.execute(
        """
        INSERT INTO documents (course, section, question, answer, embedding)
        VALUES (%s, %s, %s, %s, %s::vector)
        """,
        (doc["course"], doc["section"], doc["question"], doc["answer"],
         vec_to_str(vec))
    )
conn.commit()

We loop over the documents and insert each one with its embedding. We hand Postgres the vector as text, so the ::vector cast tells it to parse that string back into a vector. We call conn.commit() to persist the changes.

Searching with cosine similarity

Search with a query:

query = "I just discovered the course. Can I still join it?"
query_vector = model.encode(query)
query_str = vec_to_str(query_vector)

Search for the most similar documents:

results = con
## Key concepts

- [OpenAI API](/course-wiki/openai-api/)
- [RAG](/course-wiki/rag/)
- [Docker](/course-wiki/docker/)
- [PostgreSQL](/course-wiki/postgresql/)
- [Vector Search](/course-wiki/vector-search/)
- [Embeddings](/course-wiki/embeddings/)
- [Elasticsearch](/course-wiki/elasticsearch/)
- [Jupyter Notebooks](/course-wiki/jupyter-notebooks/)
## Related notes

- [llmz-m01-agents](/course-wiki/llmz-m01-agents/)
- [llmz-m01-building-the-prompt](/course-wiki/llmz-m01-building-the-prompt/)
- [llmz-m01-data-ingestion](/course-wiki/llmz-m01-data-ingestion/)
- [llmz-m01-environment](/course-wiki/llmz-m01-environment/)
- [llmz-m01-function-calling](/course-wiki/llmz-m01-function-calling/)
- [llmz-m01-introduction](/course-wiki/llmz-m01-introduction/)
- [llmz-m01-other-frameworks](/course-wiki/llmz-m01-other-frameworks/)
- [llmz-m01-quick-rag-revision-optional](/course-wiki/llmz-m01-quick-rag-revision-optional/)
- [llmz-m01-rag](/course-wiki/llmz-m01-rag/)
- [llmz-m01-rag-helper](/course-wiki/llmz-m01-rag-helper/)
- [llmz-m01-search](/course-wiki/llmz-m01-search/)
- [llmz-m01-the-agentic-loop](/course-wiki/llmz-m01-the-agentic-loop/)
- [llmz-m01-the-course-faq-dataset](/course-wiki/llmz-m01-the-course-faq-dataset/)
- [llmz-m01-the-llm](/course-wiki/llmz-m01-the-llm/)
- [llmz-m01-toyaikit](/course-wiki/llmz-m01-toyaikit/)
- [llmz-m01-wrap-up-of-part-1](/course-wiki/llmz-m01-wrap-up-of-part-1/)
- [llmz-m02-embedding-our-dataset](/course-wiki/llmz-m02-embedding-our-dataset/)
- [llmz-m02-embeddings](/course-wiki/llmz-m02-embeddings/)
- [llmz-m02-next-steps](/course-wiki/llmz-m02-next-steps/)
- [llmz-m02-rag-with-vector-search](/course-wiki/llmz-m02-rag-with-vector-search/)
- [llmz-m02-using-onnx-runtime-instead-of-pytorch](/course-wiki/llmz-m02-using-onnx-runtime-instead-of-pytorch/)
- [llmz-m02-vector-search](/course-wiki/llmz-m02-vector-search/)
- [llmz-m02-vector-search-2](/course-wiki/llmz-m02-vector-search-2/)
- [llmz-m02-vector-search-with-minsearch](/course-wiki/llmz-m02-vector-search-with-minsearch/)
- [llmz-m02-vector-search-with-sqlitesearch](/course-wiki/llmz-m02-vector-search-with-sqlitesearch/)
- [llmz-m03-ai-agents](/course-wiki/llmz-m03-ai-agents/)
- [llmz-m03-ai-copilot](/course-wiki/llmz-m03-ai-copilot/)
- [llmz-m03-ai-orchestration](/course-wiki/llmz-m03-ai-orchestration/)
- [llmz-m03-best-practices](/course-wiki/llmz-m03-best-practices/)
- [llmz-m03-context-engineering](/course-wiki/llmz-m03-context-engineering/)
- [llmz-m03-next-steps](/course-wiki/llmz-m03-next-steps/)
- [llmz-m03-retrieval-augmented-generation](/course-wiki/llmz-m03-retrieval-augmented-generation/)
- [llmz-m03-setting-up-kestra](/course-wiki/llmz-m03-setting-up-kestra/)
- [llmz-m04-agent-evaluation](/course-wiki/llmz-m04-agent-evaluation/)
- [llmz-m04-evaluation](/course-wiki/llmz-m04-evaluation/)
- [llmz-m04-generating-ground-truth-data](/course-wiki/llmz-m04-generating-ground-truth-data/)
- [llmz-m04-generating-ground-truth-for-all-documents](/course-wiki/llmz-m04-generating-ground-truth-for-all-documents/)
- [llmz-m04-generating-rag-answers](/course-wiki/llmz-m04-generating-rag-answers/)
- [llmz-m04-llm-as-a-judge](/course-wiki/llmz-m04-llm-as-a-judge/)
- [llmz-m04-next-steps](/course-wiki/llmz-m04-next-steps/)
- [llmz-m04-rag-and-agent-evaluation](/course-wiki/llmz-m04-rag-and-agent-evaluation/)
- [llmz-m04-search-evaluation](/course-wiki/llmz-m04-search-evaluation/)
- [llmz-m04-search-parameter-tuning](/course-wiki/llmz-m04-search-parameter-tuning/)
- [llmz-m05-assistant](/course-wiki/llmz-m05-assistant/)
- [llmz-m05-built-in-judge](/course-wiki/llmz-m05-built-in-judge/)
- [llmz-m05-capturing-metrics](/course-wiki/llmz-m05-capturing-metrics/)
- [llmz-m05-chat-app](/course-wiki/llmz-m05-chat-app/)
- [llmz-m05-docker-compose](/course-wiki/llmz-m05-docker-compose/)
- [llmz-m05-grafana-dashboards](/course-wiki/llmz-m05-grafana-dashboards/)
- [llmz-m05-monitoring](/course-wiki/llmz-m05-monitoring/)
- [llmz-m05-next-steps](/course-wiki/llmz-m05-next-steps/)
- [llmz-m05-querying-data](/course-wiki/llmz-m05-querying-data/)
- [llmz-m05-storing-data-in-postgresql](/course-wiki/llmz-m05-storing-data-in-postgresql/)
- [llmz-m05-streamlit-dashboard](/course-wiki/llmz-m05-streamlit-dashboard/)
- [llmz-m05-synthetic-data-generation](/course-wiki/llmz-m05-synthetic-data-generation/)
- [llmz-m05-user-feedback](/course-wiki/llmz-m05-user-feedback/)
- [llmz-m06-best-practices-for-rag](/course-wiki/llmz-m06-best-practices-for-rag/)
- [llmz-m06-document-reranking](/course-wiki/llmz-m06-document-reranking/)
- [llmz-m06-hybrid-search](/course-wiki/llmz-m06-hybrid-search/)
- [llmz-m06-hybrid-search-with-langchain](/course-wiki/llmz-m06-hybrid-search-with-langchain/)
- [llmz-m06-next-steps](/course-wiki/llmz-m06-next-steps/)
- [llmz-m07-chunking-for-longer-texts](/course-wiki/llmz-m07-chunking-for-longer-texts/)
- [llmz-m07-content-processing-cases-and-steps](/course-wiki/llmz-m07-content-processing-cases-and-steps/)
- [llmz-m07-end-to-end-project-example](/course-wiki/llmz-m07-end-to-end-project-example/)
- [llmz-m07-evaluating-rag](/course-wiki/llmz-m07-evaluating-rag/)
- [llmz-m07-evaluating-retrieval](/course-wiki/llmz-m07-evaluating-retrieval/)
- [llmz-m07-interface-and-ingestion-pipeline](/course-wiki/llmz-m07-interface-and-ingestion-pipeline/)
- [llmz-m07-monitoring-and-containerization](/course-wiki/llmz-m07-monitoring-and-containerization/)
- [llmz-m07-summary-and-closing-remarks](/course-wiki/llmz-m07-summary-and-closing-remarks/)
## Sources

- [Video](https://www.youtube.com/watch?v=0P54MFyz-mc&list=PL3MmuxUbc_hLZFNgSad56pDBKK8KO0XIv)
- [Lesson file](https://github.com/llm-zoomcamp/blob/main/llm-zoomcamp/cohorts/2026/02-vector-search/08-pgvector.md)

DataTalks.Club