Uploaded October 2025 | Updated November 2026, 3 hours ago
Do you want to turn your Postgres Database into a Vector Store?
Description:
In this tutorial, you will learn how to enable the pgvector extension to convert your Neon Postgres database into a vector database. This allows you to store and earch vector embeddings, which are necessary components when building Retrieval Augmented Generation (RAG) systems. The tutorial uses the Neon SQL editor for all the necessary steps.
Get 20% off Hostinger VPS (referral link):
https://hostinger.com?REFERRALCODE=BYS0JAIWCIJP
Wispr Flow is my tool of choice for fast, polished voice-to-text in any app.
ref.wisprflow.ai/mike-murphy (affiliate link)
Vector Store:
The database or table that holds your embeddings and metadata
Embeddings:
The numeric representation of text, image, or audio content
RAG System:
Process that searches vector store for relevant embeddings and then feeds to LLM for outputs.
pgvector:
Postgres extension that enables Postgres to store embeddings/vectors.
In this tutorial, you'll learn:
👉 SQL Editor
👉 Save SQL Queries
👉 Enable pgvector extension
👉 Verify pgvector
👉 Create a table to store embeddings
👉 Add vector index
👉 Insert test row data
👉 Similarity Search
1. Neon Account Sign-Up:
neon.com
2. The pgvector extension (Neon Documentation):
neon.com/docs/extensions/pgvector
3. How To Create Neon Postgres Database Tutorial:
youtu.be/doIFdMp7D2E
——————
✅ Chapters:
00:00 Intro: About The Tutorial
00:32 What You Will Learn
00:52 Prerequites
01:11 Tutorial: How To Setup Neon Database
01:24 Step 1: Open SQL Editor
01:35 Sign In To Neon.com
01:49 Open Projects Dashboard
01:58 Open SQL Editor
02:07 Select Database
02:29 Step 2: Enable pgvector
02:41 Neon Docs (pgvector)
03:02 IF NOT EXISTS
03:49 SQL Editor: Enable pgvector
04:22 RUN Command
04:42 Step 3: Verify pgvector is enabled
04:50 SQL Editor Basics
05:15 Add Comments in SQL Editor
05:53 How To Run Commands
06:13 Step 4: Create Embeddings Table
06:42 Create Table Code Overview
07:27 Embedding Vector (Dimension)
08:05 LLM Embedding Models
08:55 Confirm Table Was Inserted
09:14 Save Query
09:55 Step 5: Add Vector Index
10:42 Ivfflat little date warning
10:58 Step 6: Insert Test Row
11:41 Embedding Vector (4) for Demo
12:19 Documents_Demo Table
12:33 Embedding Dimension in Header
12:59 Create New Query
13:33 Confirm Test Data Row
13:50 Similarity Search
14:15 How RAG Works
15:39 Similarity Search Test
——————
🪜 How To Enable PGVECTOR Extension on Neon Postgres Database:
1. Sign into neon.com
2. Open the SQL Editor
3. Log in to your Neon dashboard
4. Select your organization
5. Click 'Projects,' open your project
6. Click the 'SQL editor' and select correct database
7. Copy/Paste CREATE EXTENSION vector IF NOT EXISTS) into the SQL editor.
8. Press 'Run' to enable pgvector
9. Verify the Extension to confirm that the pgvector extension is successfully enabled.
10. Create the Vector Store Table (documents)
11. Ensure the embedding vector dimension number (e.g., 1536) matches the dimension used by your chosen Large Language Model.
12. Add a Vector Index: Copy and run the code snippet to add a vector index to your new table, which will speed up similarity searches.
13. Test the Functionality: Insert a test row of sample data
Run a similarity search query to ensure the database can correctly make a match between text and vectors.
——————Resources———————
➜Gear I Use: mikemurphy.co/resources
➜Adobe After Effects: mikemurphy.co/adobe
➜Terrapin Textures: terrapintextures.com
➜ Adobe Audition Course (Domestika): mikemurphy.co/domestika
➜Amazon Influencer Store: amazon.com/shop/mikemurphyco
➜See All My Gear: mikemurphy.co/podgear
➜ Get Adobe Creative Cloud: mikemurphy.co/adobe
➜ Screenflow: mikemurphy.co/screenflow
➜ Website: mikemurphy.co
➜ Podcast: mikemurphy.co/mmu
➜ Instagram: instagram.com/mikeunplugged
➜ X: http://x.com/mikeunplugged
➜ YouTube: mikemurphy.co/youtube
Amazon Associates Disclosure:
Mike Murphy is a participant in the Amazon Services LLC Associates Program, an affiliate advertising program designed to provide a means for sites to earn advertising fees by advertising and linking to Amazon.com
Do you want to turn your Postgres Database into a Vector Store?
Description:
In this tutorial, you will learn how to enable the pgvector extension to convert your Neon Postgres database into a vector database. This allows you to store and earch vector embeddings, which are necessary components when building Retrieval Augmented Generation (RAG) systems. The tutorial uses the Neon SQL editor for all the necessary steps.
Get 20% off Hostinger VPS (referral link):
https://hostinger.com?REFERRALCODE=BYS0JAIWCIJP
Wispr Flow is my tool of choice for fast, polished voice-to-text in any app.
ref.wisprflow.ai/mike-murphy (affiliate link)
Vector Store:
The database or table that holds your embeddings and metadata
Embeddings:
The numeric representation of text, image, or audio content
RAG System:
Process that searches vector store for relevant embeddings and then feeds to LLM for outputs.
pgvector:
Postgres extension that enables Postgres to store embeddings/vectors.
In this tutorial, you'll learn:
👉 SQL Editor
👉 Save SQL Queries
👉 Enable pgvector extension
👉 Verify pgvector
👉 Create a table to store embeddings
👉 Add vector index
👉 Insert test row data
👉 Similarity Search
1. Neon Account Sign-Up:
neon.com
2. The pgvector extension (Neon Documentation):
neon.com/docs/extensions/pgvector
3. How To Create Neon Postgres Database Tutorial:
youtu.be/doIFdMp7D2E
——————
✅ Chapters:
00:00 Intro: About The Tutorial
00:32 What You Will Learn
00:52 Prerequites
01:11 Tutorial: How To Setup Neon Database
01:24 Step 1: Open SQL Editor
01:35 Sign In To Neon.com
01:49 Open Projects Dashboard
01:58 Open SQL Editor
02:07 Select Database
02:29 Step 2: Enable pgvector
02:41 Neon Docs (pgvector)
03:02 IF NOT EXISTS
03:49 SQL Editor: Enable pgvector
04:22 RUN Command
04:42 Step 3: Verify pgvector is enabled
04:50 SQL Editor Basics
05:15 Add Comments in SQL Editor
05:53 How To Run Commands
06:13 Step 4: Create Embeddings Table
06:42 Create Table Code Overview
07:27 Embedding Vector (Dimension)
08:05 LLM Embedding Models
08:55 Confirm Table Was Inserted
09:14 Save Query
09:55 Step 5: Add Vector Index
10:42 Ivfflat little date warning
10:58 Step 6: Insert Test Row
11:41 Embedding Vector (4) for Demo
12:19 Documents_Demo Table
12:33 Embedding Dimension in Header
12:59 Create New Query
13:33 Confirm Test Data Row
13:50 Similarity Search
14:15 How RAG Works
15:39 Similarity Search Test
——————
🪜 How To Enable PGVECTOR Extension on Neon Postgres Database:
1. Sign into neon.com
2. Open the SQL Editor
3. Log in to your Neon dashboard
4. Select your organization
5. Click 'Projects,' open your project
6. Click the 'SQL editor' and select correct database
7. Copy/Paste CREATE EXTENSION vector IF NOT EXISTS) into the SQL editor.
8. Press 'Run' to enable pgvector
9. Verify the Extension to confirm that the pgvector extension is successfully enabled.
10. Create the Vector Store Table (documents)
11. Ensure the embedding vector dimension number (e.g., 1536) matches the dimension used by your chosen Large Language Model.
12. Add a Vector Index: Copy and run the code snippet to add a vector index to your new table, which will speed up similarity searches.
13. Test the Functionality: Insert a test row of sample data
Run a similarity search query to ensure the database can correctly make a match between text and vectors.
——————Resources———————
➜Gear I Use: mikemurphy.co/resources
➜Adobe After Effects: mikemurphy.co/adobe
➜Terrapin Textures: terrapintextures.com
➜ Adobe Audition Course (Domestika): mikemurphy.co/domestika
➜Amazon Influencer Store: amazon.com/shop/mikemurphyco
➜See All My Gear: mikemurphy.co/podgear
➜ Get Adobe Creative Cloud: mikemurphy.co/adobe
➜ Screenflow: mikemurphy.co/screenflow
➜ Website: mikemurphy.co
➜ Podcast: mikemurphy.co/mmu
➜ Instagram: instagram.com/mikeunplugged
➜ X: http://x.com/mikeunplugged
➜ YouTube: mikemurphy.co/youtube
Amazon Associates Disclosure:
Mike Murphy is a participant in the Amazon Services LLC Associates Program, an affiliate advertising program designed to provide a means for sites to earn advertising fees by advertising and linking to Amazon.com










