Get running

The default deployment gives you a local Kind cluster running the platform plus monitoring — Prometheus, Grafana, the otel operator, ingress-nginx. More comprehensive examples and deployments on Kubernetes infrastructure are documented in the guide.

Partly written

Steps 1 and 2 are written. Step 3 — the first pipeline — is still a skeleton, because the seeding and scenario runners are being rebuilt.

The three steps

1. Prerequisites — five tools and a container runtime. Nothing else on this path asks you to install anything.
2. Deploy the stack — two commands: create the cluster, apply the releases.
3. Run your first pipeline — connect to your sources and destinations, create your first data pipeline, and watch change events arrive.

What this costs you

Time	~10 minutes warm, ~25 minutes on a cold image cache
Memory	8 GB RAM free (the stack does not fit comfortably in less)
Disk	~30 GB, mostly container images
Network	Pulls from quay.io, docker.io and ghcr.io

Everything runs locally and is deleted with a single teardown command. No cloud account is involved unless you follow one of the cloud guides.