AI agent for document retrieval and intelligent answering
Workflow Description
Creates an AI agent that retrieves documents from Google Drive, processes them with embeddings, stores them in a vector database, and answers user questions based on the retrieved document content using natural language understanding.
How it works
- 1.Load documents from Google Drive and automatically split their text into chunks
- 2.Convert text chunks into numerical embeddings and store them in Milvus vector database
- 3.Activate an AI agent that receives and processes questions via chat interface
- 4.Search the vector store and retrieve the most relevant document sections
- 5.Generate answers from OpenAI model based on the retrieved documents
Use cases
- Build an intelligent assistant that answers questions about company documents and policies
- Automate customer inquiry responses using a digital knowledge base
- Create a semantic search engine that understands meaning and context
Requirements
- Google Drive account with documents ready for analysis
- OpenAI and Cohere API keys for embeddings and language models
- Milvus vector database instance configured and accessible
Service Value
Ideal as a smart automation service combining integrations and AI to produce ready-to-use results.
Apps Used
Details
How to Use
- 1.Click "Download Template"
- 2.Open your n8n dashboard
- 3.Go to Workflows > Import from File
- 4.Select downloaded file and configure credentials
Nodes Used (14)
Default Data Loader
Document Default Data Loader
Embeddings Cohere
Embeddings Cohere
When chat message received
Chat Trigger
Watch New Files
Google Drive Trigger
Download New
Google Drive
Insert into Milvus
Vector Store Milvus
RAG Agent
Agent
Retrieve from Milvus
Vector Store Milvus
OpenAI 4o
OpenAI
Memory
Memory Buffer Window
Cohere embeddings
Embeddings Cohere
Set Chunks
Text Splitter Recursive Character Text Splitter
Extract from File
Extract From File
Sticky Note
Sticky Note