Vector Stores on AWS

123456789101112
-- Enable the pgvector extension
CREATE EXTENSION IF NOT EXISTS vector;

-- Create a table for document embeddings
CREATE TABLE document_embeddings (
    id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
    document_id VARCHAR(255) NOT NULL,
    chunk_index INTEGER NOT NULL,
    content TEXT NOT NULL,
    embedding vector(1536),  -- OpenAI text-embedding-3-small dimension
    metadata JSONB DEFAULT '{}',
    tenant_id VARCHAR(100) NOT NULL,

123456789101112
{
  "size": 10,
  "query": {
    "hybrid": {
      "queries": [
        {
          "match": {
            "content": {
              "query": "kubernetes pod scheduling",
              "boost": 0.3
            }
          }

123456789101112
from opensearchpy import OpenSearch, RequestsHttpConnection
from requests_aws4auth import AWS4Auth
import boto3

# Initialize client with IAM authentication
region = 'us-east-1'
service = 'aoss'
credentials = boto3.Session().get_credentials()
awsauth = AWS4Auth(credentials.access_key, credentials.secret_key,
                   region, service, session_token=credentials.token)

client = OpenSearch(

123456789101112
import psycopg2
from pgvector.psycopg2 import register_vector
import numpy as np

# Connection with pgvector extension
conn = psycopg2.connect(
    host="your-aurora-cluster.cluster-xxx.us-east-1.rds.amazonaws.com",
    database="vectors_db",
    user="admin",
    password="your-password"
)
register_vector(conn)

123456789101112
import boto3
import json
from opensearchpy import OpenSearch, RequestsHttpConnection
from requests_aws4auth import AWS4Auth

class VectorSearchService:
    def __init__(self, collection_endpoint: str, index_name: str):
        self.region = 'us-east-1'
        self.service = 'aoss'
        credentials = boto3.Session().get_credentials()
        self.awsauth = AWS4Auth(
            credentials.access_key,

123456789101112
from typing import List, Dict
import asyncio
import asyncpg

class HybridSearchService:
    def __init__(self, connection_string: str):
        self.conn_string = connection_string
        self.pool = None
    
    async def initialize(self):
        self.pool = await asyncpg.create_pool(self.conn_string, min_size=5, max_size=20)
    

123456789101112
import numpy as np
from typing import List, Tuple

def maximal_marginal_relevance(
    query_embedding: np.ndarray,
    candidate_embeddings: np.ndarray,
    candidate_contents: List[str],
    k: int = 5,
    lambda_param: float = 0.7,
    initial_k: int = 20
) -> List[Tuple[int, str, float]]:
    """