New to Bundata? Use this hub to learn core concepts, run your first extraction, design a schema, run vector search, build an agent, and orchestrate workflows.
Get started
New to Bundata? Start here to learn core concepts and run your first extraction, schema, vector search, agent, and workflow. This section is your entry point to the document intelligence layer.
Overview
Bundata turns unstructured documents into vector-ready intelligence through context-aware and schema-aware extraction. You define what to extract, connect your sources, run extraction to produce smart bites, then use the Vector Catalog and Vector Search for RAG, agents, and workflows.
Follow the paths below: begin with the quickstart or core concepts, then move to your first extraction, schema, vector search, agent, or workflow.
Start here
Beginner tutorials
Step-by-step guides to run your first workflows.
Create your first extraction
Run context-aware extraction and get smart bites.
Build your first schema
Design a schema in Schema Studio.
Run your first vector search
Query the Vector Catalog with natural language.
Build your first agent
Ground an agent with the Vector Catalog.
Run your first workflow
Orchestrate extraction and downstream steps.
Deployment model
Cloud and self-hosted options.
Core concepts
- Core concepts — Accounts, workspaces, authentication, connectors, schemas, pipelines, Vector Catalog
- Deployment model — Bundata Cloud, AWS, Azure, GCP, self-hosted