AI-Powered RAG Document Processing Chatbot with Google Drive Supabase OpenAI

## 📋 Data Processing & RAG Workflow

### 📁 Step 1: Google Drive Trigger  
Watches a specific Google Drive folder for new file uploads. Triggers every minute.

### 🔁 Step 2: Loop Over Uploaded Files  
Ensures each uploaded file is processed one-by-one to avoid bulk processing issues.

### 🆔 Step 3: Set File ID  
Extracts and stores the Google Drive file ID for reference in later steps.

### 📥 Step 4: Download File  
Downloads the file (PDF, CSV, or Google Docs). Google Docs are auto-converted to PDF.

### 🧭 Step 5: File Type Switch  
Routes file based on type:
- PDF → Go to PDF Text Extractor
- CSV → Go to CSV Extractor

### 📄 Step 6A: Extract Text from PDF  
Extracts raw text content from the PDF file.

### 📊 Step 6B: Extract Text from CSV  
Extracts and formats CSV data into a readable string.

### 📦 Step 7: Format Document Data  
Wraps extracted text into JSON: `{"data": "text content"}`.

### 🧠 Step 8: Generate Metadata (Gemini)  
Uses Gemini AI to create a document title and description.

### ✂️ Step 9: Split into Chunks  
Splits the document text into 1000-character chunks with 200-character overlap, intelligently at paragraph/sentence/word level.

### 🔄 Step 10: Split Out Chunks  
Converts array of chunks into individual workflow items.

### 🚫 Step 11: Limit to 20 Chunks  
Limits processing to first 20 chunks to control API and processing load.

### 🧠 Step 12: Enhance Chunk Context (Gemini)  
Each chunk is enhanced for better clarity, continuity, and context using Gemini.

### 🧩 Step 13: Summarize  
Combines enhanced chunks using `###SPLIT###` separator for further processing.

### 🧷 Step 14: Re-Split Enhanced Chunks  
Splits combined text back into chunks using `###SPLIT###` separator.

### 🧬 Step 15: Generate Embeddings (OpenAI)  
Converts text chunks into vector embeddings for semantic search.

### 🗂️ Step 16: Add Metadata  
Each chunk is wrapped with metadata (file ID, title, description) for vector DB.

### 🧠 Step 17: Store in Supabase Vector DB  
Stores embeddings in Supabase vector store for later semantic retrieval.