Core concepts

Understand how Bundata is organized: accounts, workspaces, authentication, data flow, and pipeline concepts.

Accounts and workspaces

A Bundata account is the top-level entity for billing and support. A workspace is your environment for running pipelines, managing connectors, and accessing the Vector Catalog. You can have multiple workspaces per account. Users are invited to workspaces with configurable access.

Authentication and access

API key — Use an API key to authenticate with the Bundata API. Create and manage keys in the dashboard. Use in requests as Authorization: Bearer <key>.
User identity — Tied to your account (e.g. email). Roles control access to workspaces, connectors, and pipelines.

Bundata interfaces

Platform (UI) — Design and run pipelines without code: Extraction Studio, Workflows, Schema Studio.
API — Programmatic access for partitioning, enrichment, chunking, and embedding. See API Reference.
CLI / SDKs — Where available, for scripting and integration.

Data flow and pipeline concepts

Connectors — Source connectors pull documents from storage and apps; destination connectors send smart bites to vector stores or object storage. See Integrations overview.
Schema — Defines the structure of your output (elements, metadata, chunk shape). Defined in Schema Studio or via API.
Pipeline — End-to-end path: ingest → partition → enrich → chunk → (optionally) embed → deliver. Workflows in the UI schedule and run these steps.
Vector Catalog — Managed store for embeddings and indexed smart bites; supports semantic search and agent grounding.

Billing and usage

Billing is at the account level. Usage is often measured in pages processed. See API Reference — Pricing and Sign up.

Next steps

Sign up — Create an account and choose a plan.
First extraction — Run your first pipeline.
Integrations overview — Connectors and cloud deployments.