Chat with PDF, CSV, and JSON documents using Google Gemini RAG

Go to Workflow
0 views
Built by Md Khalid Ali Md Khalid Ali
Created on June 20, 2026

Description

Overview

Turn documents into an AI-powered knowledge base.

Upload PDF, CSV, or JSON files and ask natural-language questions about their content using a Retrieval-Augmented Generation (RAG) workflow powered by Google Gemini. The workflow extracts, embeds, and semantically searches document data to generate accurate, source-grounded answers.

Designed as a simple and extensible starting point for building AI document assistants.

Key Features

Upload and analyze PDF, CSV, and JSON
AI chatbot with semantic document search
Retrieval-Augmented Generation (RAG) architecture
Answers grounded in uploaded documents
Beginner-friendly workflow with clear documentation
Easy to extend for production use

How It Works

Upload a document via form trigger
Content is split into searchable chunks
Gemini generates embeddings
Data is stored in a vector store
The chatbot retrieves context and answers questions

Requirements

Google Gemini API credentials

Notes

Uses an in-memory vector store (data resets on restart)
Can be replaced with Pinecone, Supabase, Weaviate, or other persistent databases
Gemini API usage may incur costs depending on document size and query volume

Nodes Used (7)

AI Agent
@n8n/n8n-nodes-langchain.agent
Default Data Loader
@n8n/n8n-nodes-langchain.documentDefaultDataLoader
Embeddings Google Gemini
@n8n/n8n-nodes-langchain.embeddingsGoogleGemini
Google Gemini Chat Model
@n8n/n8n-nodes-langchain.lmChatGoogleGemini
Recursive Character Text Splitter
@n8n/n8n-nodes-langchain.textSplitterRecursiveCharacterTextSplitter
Simple Memory
@n8n/n8n-nodes-langchain.memoryBufferWindow
Simple Vector Store
@n8n/n8n-nodes-langchain.vectorStoreInMemory