Skip to main content
Think of pragma-os like iOS for your data infrastructure. Instead of manually wiring together storage, compute, and AI services, you browse a store of pre-built resources and install what you need. Dependencies resolve automatically. When upstream data changes, downstream resources update themselves.

​
How It Works

1

Browse

Explore the store for the resources you need - storage buckets, datasets, vector databases, RAG pipelines, and more.
2

Configure

Set your parameters: project, region, naming. Each resource knows what it needs.
3

Auto-sync

pragma-os provisions everything and keeps it in sync. When sources change, dependent resources automatically update.

​
Why pragma-os

Install in minutes, not weeks

Skip the 2-3 weeks of manual setup. Get production-ready infrastructure in 15 minutes.

Changes propagate automatically

Update a source schema and watch dependent resources adapt. No manual intervention.

No platform team required

Stop waiting on infrastructure tickets. Install what you need, when you need it.

​
Example: RAG Pipeline

Building a RAG pipeline traditionally means weeks of work - setting up storage, configuring vector databases, wiring embedding services, managing sync jobs. With pragma-os, you describe what you want: 
provider: gcp
resource: storage
name: documents
config:
  location: EU
  lifecycle:
    delete_after_days: 365

provider: gcp
resource: bigquery-dataset
name: embeddings
config:
  location: EU
depends_on:
  - documents

pragma resources apply .
pragma-os provisions both resources and establishes the dependency. When new documents land in storage, the embeddings dataset knows about it. The full RAG pipeline example with vector search is coming soon.

​
Get Started

Quickstart

Install your first resource in 5 minutes.

How It Works

Understand the pragma-os model.

CLI Reference

Complete command reference.

API Reference

REST API documentation.