Skip to main content

Quick Setup

Get from zero to a running workspace in under 10 minutes.

Prerequisites​

  • Docker — installed automatically if missing
  • Hardware — 4+ CPU cores, 8+ GB RAM, 20+ GB free disk
  • Network — outbound access to *.docker.io, *.tracebloc.io

1. Create an Account​

Sign up at ai.tracebloc.io. Free to get started.

2. Register a Client​

Open the client page and click "+". Fill in a name, location, and password. Note your Client ID and password — you need both in the next step.

tip

Provisioning takes a few seconds. Refresh the page if the status doesn't update.

3. Deploy​

bash <(curl -fsSL https://tracebloc.io/install.sh)

The installer prompts for your workspace name, Client ID, and password. It handles Docker, Kubernetes, and the tracebloc client. Safe to re-run.

4. Verify​

kubectl get pods -n <workspace>

Two pods in Running state: mysql-... and tracebloc-jobs-manager-...

Open ai.tracebloc.io — your client should show Online.

Stuck?

Check the Troubleshooting page.

What You Now Have​

  • A local Kubernetes cluster running in Docker
  • The tracebloc client connected to the platform
  • A workspace ready for use cases

Your data never leaves your infrastructure. Only metadata and metrics are shared with the platform.

What's Next​

Data owners: Create a use case — select datasets, define metrics, invite contributors.

Data scientists: Join a use case — train and submit models on shared infrastructure.

Go Deeper​

Need custom cluster settings, GPU configuration, or manual Helm deployment? See the full Setup Guide.