Your server. Your identity.
Your data.

A dedicated machine, or your own hardware, running the whole platform under your domain and your identity provider. Departments carry their own limits, and every action by a person or an agent is on record.

Request for free trial See demo

How the trial works: sign in, tell us about your institution, and once we approve it you get a dedicated 4 vCPU / 8 GB server under your own brand, free for 30 days. Afterwards the same server carries on as the Nano plan at €30.00 a month, and you can move to a bigger machine whenever you want. Want us to set it up for your team? Done-for-you setup →

The Problem

Your teams need machines. Your tools are a mess.

Most companies stitch developer environments together from separate vendors — a cloud account per team, a hosted database here, a tunnel there, and a different login for each — then pay again for every person who joins. The data ends up on someone else’s servers, and nobody can say who changed what.

  • A separate vendor for machines, databases and routing
  • One platform: workspaces, six database engines, routing and TLS
  • Per-seat pricing that scales costs as you grow
  • Billed by the machine — no per-seat fee
  • Engineers waiting days for an environment
  • A workspace from a template, serving in minutes
  • A different login for every tool
  • Your own identity provider, on everything
  • Nobody can say who changed what
  • Every action by a person or an agent on record
  • Your data on someone else’s servers
  • Your hardware, or a dedicated machine of ours

Platform Architecture

Everything your organization needs. One stack.

Connectivity, workspaces and governance — delivered as one platform. No separate vendor contracts, no integration projects.

Connectivity

  • A private network per person a WireGuard address, from the office or from home
  • Browser-based workspaces nothing to install on the laptop
  • Role-based access per person and per department, enforced not reported
  • Projects on real domains a certificate on every route
  • A dedicated public IP on every plan
  • Always on a workspace runs whether or not anyone is connected

Workspaces & Services

  • One machine, many departments no hardware per employee
  • 114+ templates in the store fork one, or publish your own
  • Security ranges Kali and CTF machines, isolated per department
  • Six managed engines MySQL, MariaDB, PostgreSQL, MongoDB, RabbitMQ and Valkey (Redis-compatible)
  • Embedded and IoT templates Zephyr RTOS, Buildroot, U-Boot, ESP32
  • Workflow automation templates n8n and Node-RED

Governance & Audit

  • Your identity provider GitLab, Google, Entra or any OIDC issuer
  • The OAuth application is yours nobody signs in through us
  • Departments with ceilings members, storage, workspace and lab quotas, enforced rather than reported
  • Audit by default MCP calls, deploys, transfers and admin actions, all recorded
  • Data export and erasure built in, not a support ticket
  • Agents under the same rules an agent holds the permissions of the person who connected it

On-Premise

Run the whole platform on your own servers and network. Best when the data has to stay inside the building.

Managed Deployment

A dedicated machine on bare metal we own, serving 165 seconds after the order. Nothing for your IT team to rack.

Hybrid Cloud

Sensitive work on your own hardware, the rest on a dedicated machine of ours, each running the full platform.

AI Infrastructure

Connect Your Cloud. Empower Your Agents.

Connect the agents your engineers already use. They work on the same platform, with the same permissions and the same audit trail as the people who connected them.

PROTOCOL

One MCP Connection

Claude Code, Codex or any MCP client, working under the permissions of the person who connected it.

GUARDRAILS

Tools Scoped By Role

A model sees only the tools its user’s role allows, and every call goes through five checks before it runs.

STORAGE

Vector Store

Qdrant is a template in the store: deploy it beside the workspace that needs it, on your own machine.

MONITORING

Every Call Audited

Allowed or refused, each call is written to the account and the client that made it, per department.

Engineering Impact

Accelerate Engineering Capability

Engineers spend their time on the work, not on the machine. Environments are ready before anyone asks, a problem can be reopened exactly as it happened, and the platform grows with the company.

  • Hands-On Engineering Practice

    Real Linux machines for building, testing and breaking things, kept well away from production.

  • Faster Problem Solving

    A workspace outlives the incident, so the state that produced a bug can be opened again and looked at.

  • Team Capability at Scale

    Departments, templates and limits grow with the company, on one platform rather than one per team.

What You Get

One organization, a department per team

  1. 01

    Organization

    One root: your name, your domain and your identity provider. Everything else is set up beneath it.

  2. 02

    Departments

    Departments keep teams apart, each with its own admins, members and limits. Each can see only its own.

  3. 03

    Templates

    Every workspace starts from a template. Shared language, shared tooling, shared config. Engineers clone a template, not a wiki page.

  4. 04

    Workspace

    A workspace is a running, personalized environment. Health checks, routing, and certificates are wired before the engineer opens a browser.

Compatibility

The same platform, for Claude and Codex

An agent runs in the same environment the engineer works in - same directories, same tools, same login. What it changes shows up where the engineer can see it.

  • VS Code
  • Cursor
  • Windsurf
  • Zed
  • Neovim
  • Helix
  • Claude
  • Codex
  • Gemini CLI
  • GitHub Copilot
  • Continue
  • Aider
  • Docker
  • Nix
  • Jupyter
  • tmux
  • Wezterm
  • Ghostty

Use Cases

What teams do with it

MULTI-TEAM PLATFORM

One machine, many departments

Sales, eng, design — each department is isolated but on the same box.

INFRA COST

One bill, per machine

Development environments move off a sprawl of cloud VMs onto machines you size, billed by the hour.

TEMPLATE LIBRARY

Standardized stacks

Publish internal templates. Teams clone them, not your wiki page.

Business Impact

What leadership sees on the other side.

Not tool names: how fast a team gets going, what the bill does as the company grows, and whether an auditor gets an answer.

165 s click to a serving platform

Live the same day

A dedicated platform is serving 165 seconds after the order. No procurement quarter, no rack to wait for.

0 per-seat fees

Headcount does not move the bill

You pay for the machines. Onboarding the hundredth engineer costs the same as the tenth.

114+ templates in the store

Standard stacks, not wiki pages

Publish the internal stack once and every team starts from it, versioned and restorable.

6 managed database engines

Databases without a ticket queue

MySQL, MariaDB, PostgreSQL, MongoDB, RabbitMQ and Valkey, with credentials issued per person.

You pay for machines, not for people.

One platform replaces the cloud account per team, the hosted database, the tunnel and the extra login. It is billed by the machine and by the hour, so the bill follows the work you run rather than the size of the payroll.

Per-seat fee
None
Billed
Per machine, by the hour
Your own hardware
Supported

Start small.
Prove it in one quarter.

Start with one team on one machine. Once it is working — environments ready on day one, projects on real URLs, every action on record — add departments across the organization, with no per-seat fee.

Email Us Directly

support@selfmade.ninja selfmade.ninja labs.selfmade.ninja