New VDF AI Enterprise is now available on AWS Marketplace
Free learning path 5

Self-Hosting Basics

This free course teaches the basics of self-hosting VDF AI. You learn how the portal, packages and short-lived pull credentials work, and how to plan compute, storage, networking, identity and governance for a pilot and for production. Then you fit TLS, single sign-on and secrets to your edge, and validate and operate a deployment.

  • 4 lessons
  • 36 min
  • Beginner
  • Free, self-paced
Start lesson 1 → Open VDF AI free
Outcomes

What you will be able to do

  • Explain what the portal does, what it leaves to you and why pull credentials are low-risk
  • Choose pilot and production sizing for hosts, storage and PostgreSQL
  • Fit image pulls, ingress, TLS, single sign-on and secrets around your existing edge
  • Bring package images into an air-gapped network through your own registry
  • Validate a first deployment and hand over a runbook for updates, backups and monitoring
Before you start

Who it is for and what you need

Platform, infrastructure and SRE engineers planning a first self-hosted deployment, and the compliance and security leads who review it.

  • Working knowledge of Linux hosts and containers
  • Nothing else to read the lessons; to deploy, an active portal.vdf.ai account and a host or cluster you may use
The lessons

4 lessons, in order

Each lesson is an article grounded in the product docs. For every field and option, see the Self-hosting documentation.

  1. 1 How Self-Hosting VDF AI Works: Portal and Credentials How self-hosted VDF AI reaches your environment: the portal, signed container packages, short-lived read-only pull credentials and a one-way trust model. Guide
    8 min
  2. 2 How to Plan a Self-Hosted AI Deployment in Your Data Centre Plan a self-hosted VDF AI deployment layer by layer: compute, storage and database, networking, identity and governance, for a pilot and for production. Guide
    10 min
  3. 3 Network, TLS, SSO and Secrets for Self-Hosted AI Fit self-hosted VDF AI into your edge: outbound-only image pulls, narrow internal traffic, TLS and SSO at your reverse proxy, secrets and the air-gapped route. Guide
    9 min
  4. 4 How to Validate and Operate a Self-Hosted AI Deployment Take self-hosted VDF AI from the pre-deploy checklist to a validated first run, then set the routine for updates, credentials, backups, logs and metrics. Guide
    9 min

Take it to production

A pilot on one host proves the package runs. Production adds sizing across nodes, your identity provider, your secrets manager and an operating routine your security team can inspect.

Book an architecture call Self-hosted VDF AI
FAQ

Frequently asked questions

Do I need a portal account to follow this path?

Not to read it. To deploy a package you need an active portal.vdf.ai account. Creating one is free, and approval is a manual review that the docs expect to take less than one business day, so register while you plan the environment.

Which VDF AI packages can I self-host?

At the time of writing, the portal catalogue lists VDF AI Compliance Foundation, an EU AI Act compliance platform. The docs say more packages will follow and will appear in your catalogue automatically once your account has access to them.

Does a self-hosted deployment need a connection to VDF AI?

Only to pull images, over outbound HTTPS to the package registry. The portal never connects to your environment, deployed packages send it no telemetry and keep running if it is unreachable. An air-gapped site receives images through a connected staging host and your internal registry.

After this path

Keep going

Try another free path, or the course that follows this one. It is free for VDF AI customers and partners.

Instructor-led · Advanced

Self-Hosting and Operations

Deploy and operate VDF AI on your own infrastructure.

4 live half-days 6 modules + capstone VDF AI Certified Platform Engineer
Free path

VDF AI Fundamentals

Start here: the platform, chat and answers you can check.

3 lessons 1 h