Overview
Every AI-native application starts with data. But data is scattered—stored across platforms, in different formats, and often unstructured. Collecting and preparing this data is the first step in building anything AI-ready.
Data ingestion is how you bring this data together. Unlike traditional ETL (Extract, Transform, Load) systems, Unbody’s ingestion is designed to handle modern AI needs. It’s agnostic to data location, format, and structure, meaning your data can come from anywhere, in any form.
Why Data Ingestion Matters
Your AI is only as good as the data it’s built on. Data ingestion unifies fragmented sources, makes raw data usable, and ensures your app has the input it needs to perform.
Two Ways to Ingest Data
1. Prebuilt Integrations
Unbody provides ready-to-use connectors for popular platforms. These save you time by making it simple to pull data without custom code.
Current integrations include:
- Google Drive: Ingest files and documents. Learn more
- Discord: Import messages and content. Learn more
- GitHub: Extract repositories and related data. Learn more
New integrations are added regularly.
2. Custom Ingestion with Push API
For everything else, there’s the Push API. Use this when you need to:
- Define custom schemas for your data.
- Ingest data in any format, structured or unstructured.
The Push API is flexible and works with non-standard or unique data sources. Learn more about the Push API.