What if you could have complete control over your email, your photos, your documents, your everything — without needing to understand Linux, Kubernetes, or networking? What if self-hosting was as easy as paying a monthly subscription, but instead of feeding your data to Big Tech, you actually owned it?

This is the future I’m building towards.

The Problem: We’ve Lost Control

Every day, billions of people hand over their most intimate data to companies whose business model depends on exploiting it. Your emails, your photos, your location history, your browsing habits — all stored on servers you don’t control, governed by terms of service you didn’t read, in jurisdictions that may not protect your rights.

The technical community knows there’s an alternative. Self-hosting. Running your own services on your own hardware. But here’s the uncomfortable truth: self-hosting is currently a privilege reserved for those with technical expertise and time to spare.

I’ve spent years building and maintaining my own infrastructure. I run my own email, my own cloud storage, my own everything. And I love it. But I also recognize that expecting my non-technical friends and family to do the same is unrealistic. The barrier to entry is simply too high.

The Vision: Digital Agency for All

Here’s what I believe should exist:

You pay a monthly fee — let’s say €15-30. In return, you get:

  • Your own private cloud (photos, documents, notes)
  • Your own email server
  • Your own password manager
  • Your own calendar and contacts
  • Whatever else you need

The key differences from existing cloud providers:

  • Your data stays yours. No scanning, no profiling, no selling.
  • Runs on commodity hardware. That old PC in your closet? Perfect.
  • You can leave anytime. Take your data and go. No lock-in.
  • Transparent operations. Open source, auditable, trustworthy.

This isn’t a new idea. Projects like Nextcloud, Mailcow, and countless others already provide the building blocks. The missing piece is making it accessible to everyone — not just the technically inclined.

How It Works: The Service Cluster Model

I’ve already started building this. Here’s the architecture:

flowchart TD
    subgraph service["Service Cluster<br/>(Managed by me, runs management plane)"]
        GitOps["GitOps<br/>(ArgoCD)"]
        Monitoring["Monitoring<br/>(Grafana)"]
        Updates["Updates<br/>(Auto)"]
    end
    service --> A
    service --> B
    service --> C
    subgraph A["Customer Cluster A (Home)"]
        A1["Nextcloud"]
        A2["Vaultwarden"]
        A3["Photos"]
    end
    subgraph B["Customer Cluster B (Office)"]
        B1["Nextcloud"]
        B2["GitLab"]
        B3["CI/CD"]
    end
    subgraph C["Customer Cluster C (Colo)"]
        C1["Email"]
        C2["Matrix"]
        C3["Nextcloud"]
    end

A central service cluster manages multiple customer clusters. Each customer cluster runs on hardware owned by the customer — whether that’s a mini PC at home, a server in an office, or rented hardware in a datacenter.

The service cluster handles:

  • Deployment — GitOps-driven, declarative, reproducible
  • Monitoring — Know when something breaks before the customer notices
  • Updates — Automatic, tested, rolled back if problems occur
  • Backups — Encrypted, off-site, regularly tested

The customer cluster handles:

  • Running the actual services — Email, cloud storage, whatever they need
  • Storing the data — On hardware they physically control

This separation is crucial. The service cluster never sees customer data. It only manages the infrastructure that runs it.

Running on Old Hardware

One of my core principles: this should run on hardware people already have.

That old laptop? Those retired office PCs? That Raspberry Pi collecting dust? All viable. Kubernetes distributions like K3s run comfortably on hardware that Big Tech would consider obsolete.

This matters for several reasons:

  • Environmental — Reusing hardware instead of buying new
  • Economic — Lower barrier to entry
  • Sovereignty — True ownership means physical possession

I’ve written before about sovereign infrastructure and why physical control matters. This vision is the logical extension of that philosophy — making sovereignty accessible to everyone.

The Future: LLM-Managed Infrastructure

Here’s where it gets interesting.

Right now, I manage the service cluster. I respond to alerts, I debug issues, I make decisions. But increasingly, I’m asking: what if the service cluster could manage itself?

Large Language Models are getting remarkably good at:

  • Interpreting logs and metrics
  • Diagnosing common issues
  • Suggesting (and implementing) fixes
  • Making nuanced decisions based on context

Imagine an LLM that:

  • Notices a customer’s disk is filling up
  • Analyzes usage patterns
  • Recommends cleanup actions or storage expansion
  • Implements the chosen solution
  • Learns from the outcome

This isn’t science fiction. The building blocks exist today. What’s needed is careful integration and guardrails to ensure the LLM helps rather than harms.

The end state? Infrastructure that manages itself. My role shifts from operator to architect — designing systems and policies rather than handling day-to-day operations. The service becomes truly scalable because adding customers doesn’t linearly increase my workload.

The Business Model

Let’s be pragmatic. This vision needs to be sustainable.

For individuals: €15-20/month for basic services (cloud storage, password manager, photos).

For families: €25-35/month for shared services plus individual accounts.

For small businesses: €50-100/month for business-grade services (email, collaboration, CI/CD).

Compared to:

  • Google Workspace: €5-18/user/month (but they mine your data)
  • Microsoft 365: €5-20/user/month (same story)
  • Self-hosting yourself: “Free” but hundreds of hours of your time

The value proposition isn’t just privacy — it’s sovereignty plus convenience. You get the benefits of self-hosting without the operational burden.

What’s Next

I’m building this incrementally:

  1. Now: Running my own service cluster managing a handful of customer clusters (friends, family, early adopters)
  2. Soon: Documenting the architecture, open-sourcing the management tooling
  3. Later: Building LLM integration for autonomous operations
  4. Eventually: Making this accessible to anyone who wants true digital agency

If this vision resonates with you — whether you want to be an early adopter, contribute to the tooling, or just follow along — I’ll be documenting the journey on this blog.

The future of personal computing doesn’t have to be renting access to your own data from corporations. It can be something better. Something you actually own.

Let’s build it.