> ## Documentation Index
> Fetch the complete documentation index at: https://docs.zynfo.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Understanding AI Data Chunking and Vector Search

> Learn how ZynfoAI uses data chunking and vector embeddings to optimize your knowledge base for highly accurate AI agent responses.

# Understanding Chunking

When you ingest a large document or a massive webpage, ZynfoAI doesn't just read it as one massive block of text. Instead, it systematically breaks the text down into smaller, highly optimized pieces called **Chunks**.

## Why Chunking Matters for AI Agents

Chunking is the foundation of Retrieval-Augmented Generation (RAG). It allows the AI agent to perform highly accurate **Vector Searches**. When a user asks a question, the AI agent mathematically searches through all your chunks to find the 3-5 chunks that are most contextually similar to the user's question. It then reads only those specific chunks to formulate its exact answer.

## Viewing and Auditing Chunks

You can transparently inspect exactly how ZynfoAI has broken down your documents:

1. Go to the **Knowledge** page.
2. Find a specific source and click the **three-dot menu** (More Options).
3. Select **View Chunks**.

You will be taken to the `/knowledge/chunks` dashboard where you can read every individual paragraph or sentence segment that the AI agent has memorized from that specific source. This transparency is incredibly helpful for debugging why an AI agent might be giving a slightly inaccurate answer.

## Related Topics

* [Managing Knowledge Sources for Your AI Agent](/knowledge/managing-sources)
* [AI Agent Ingestion Methods: Websites, Docs & Drive](/knowledge/ingestion-methods)
* [Resync and Delete AI Knowledge Sources](/knowledge/resync-and-delete)