Skip to main content
Getting Started

Go From Zero to Deployed Pipeline in Under 5 Minutes

Go from zero to your first deployed agentic data pipeline in under 5 minutes.

STEP 1

Verify Cluster Connectivity

Confirm your local environment can communicate with the live ProAI cluster endpoint.

curl -X GET https://proai-server-924327084517.us-central1.run.app/health
STEP 2

Install the ProAI CLI

Download and install the official command-line interface for Mac, Linux, or Windows.

npm install -g @pro8i/proai-cli
# or with brew
brew install pro8i-platform/tap/proai
STEP 3

Initialize a New Pipeline Project

Scaffold a production pipeline template with pre-configured source, transform, and sink nodes.

proai init my-first-pipeline --template=starter-etl
cd my-first-pipeline
STEP 4

Test Locally & Deploy to Cloud

Run a local dry-run simulation with sample records, then deploy directly to production.

proai simulate --sample=100
proai deploy --env=production

Prefer clicking instead of typing? Run these same calls from the browser.

Recommended Next Steps

Where to Go After Your First Pipeline

Connect a Real Data Source

Wire PostgreSQL, Snowflake, or Kafka into your pipeline with our connector guides.

Read the guide →

Add an Agentic Flow Node

Enable autonomous schema healing and LLM-based anomaly triage on live data.

Read the guide →

Automate Deployment in CI/CD

Promote pipeline definitions between environments with the ProAI CLI and GitHub Actions.

Read the guide →

Continue From Here

Explore the documentation library or try the interactive tutorials next.