AI Agents

Product: Idea to Shop Vector - Dimension 768

8 nodes 223 153 Automatic trigger
Download

Workflow Description

An intelligent automation template that transforms product ideas into searchable vector embeddings with 768-dimensional precision, combining text segmentation, AI summarization, and vector storage for rapid semantic discovery and evaluation.

How it works

  1. 1.Trigger workflow when a new product idea is created in Notion
  2. 2.Split product descriptions into logical segments using token-aware text splitting
  3. 3.Load and validate product data from attached documents and sources
  4. 4.Generate intelligent summaries of key content sections using Gemini AI
  5. 5.Convert segmented text into 768-dimensional vectors using Gemini embeddings
  6. 6.Store vectors and metadata in Pinecone for fast semantic search and retrieval

Use cases

  • Build a searchable product idea library with instant semantic discovery capabilities
  • Identify duplicate or similar ideas automatically before development begins
  • Enable teams to find relevant historical concepts for current projects quickly

Requirements

  • Active Notion workspace with a products database configured
  • Valid Google Gemini API credentials with embedding model access
  • Configured Pinecone vector database instance with 768-dimensional indexes

Service Value

Ideal as a smart automation service combining integrations and AI to produce ready-to-use results.

Apps Used

Text Splitter Notion Document Default Data Loader Gemini Vector Store

Details

Trigger Automatic trigger
Nodes 8
Apps 5
Views 223
Downloads 153

How to Use

  1. 1.Click "Download Template"
  2. 2.Open your n8n dashboard
  3. 3.Go to Workflows > Import from File
  4. 4.Select downloaded file and configure credentials

Nodes Used (8)

Token Splitter

Text Splitter Token Splitter

#1

Notion - Page Added Trigger

Notion

#2

Notion - Retrieve Page Content

Notion

#3

Filter Non-Text Content

Filter

#4

Summarize - Concatenate Notion's blocks content

Summarize

#5

Create metadata and load content

Document Default Data Loader

#6

Embeddings Google Gemini

Gemini Model

#7

Pinecone Vector Store

Vector Store Pinecone

#8