MLflow Tracking Server

MLflow Tracking Server — MLflow tracking + model registry on PostgreSQL with basic-auth enabled. Unique admin password generated on first boot.

Deployed as a VM in your own Azure subscription — your data never leaves your tenant. The image is hardened and self-provisioning: on first boot it generates unique credentials (written to a root-only file on the VM), so no two deployments ever share a secret.

What's included

• Experiments + model registry
• VNet-first security posture
• PostgreSQL backend
• Local artifact store

How it works

• Deploy from Azure Marketplace into your subscription (any region, your VNet rules).
• SSH in and read the generated credentials file in /root/.
• No external runtime dependency on the publisher — everything runs in your tenant.

Support: contact@infoinlet.com