Heidelberg AI News

Video · David Ondrej ·

The Ultimate Linux Tutorial for AI Agents (watch before 2027)

In short

The video explains why Linux is well suited to AI agents: its text-based, scriptable, open-source environment lets agents operate servers and software through shell commands. It demonstrates using an AI agent to secure a Linux VPS and deploy a full-stack MVP with frontend, backend, PostgreSQL, ClickHouse, and Caddy, arguing that one VPS can reduce platform costs and increase control.

Key points

  • Linux powers 100% of the world’s 500 largest data centers and supercomputers, plus most cloud servers and Android devices.
  • The Linux kernel manages hardware, while shells such as Bash provide command-line control through commands like ls, cd, and pwd.
  • Ubuntu is presented as the default choice for servers, cloud workloads, VPSs, CI, Docker, and GPU machines.
  • Omarchy, based on Arch Linux and Hyperland, is presented as an agent-friendly desktop distribution with text-configured settings.
  • An AI agent secured a VPS in about 2.5 minutes by enabling updates and a firewall, restricting SSH and web traffic, disabling password login, and adding brute-force protection.
  • A full-stack massage-booking MVP ran four services on one VPS: Caddy, a Node API, PostgreSQL, and ClickHouse.
  • Fifteen isolated cloud sub-agents could run in parallel without consuming the user’s local computer resources.

Written by an AI model from the video’s captions and description.

In this video, I walk through what Linux is and why AI Agents love it. I also show how you can get your own VPS running Linux and use Cloudroom with its Linux machines for development. David Ondrej is a serial entrepreneur, who comes from a small village in Czech Republic.

Watch on YouTube The opening of the article at youtube.com.

In this story

A tool or model opens in the Heidelberg AI Curriculum: what it does, what it costs, how to start. A repo opens on GitHub.

More from David Ondrej

All stories

VideoDavid Ondrej

Build your own open-source Grok Bot, here’s how

The video demonstrates building an open-source, self-hosted alternative to Grok Bot by reverse-engineering its UI and architecture with AI coding agents. It shows how multi-agent workflows, Rust, cloud sandboxes, and VPS deployment can accelerate development while keeping humans responsible for product and architecture decisions.

VideoDavid Ondrej

My $1,000 Fable 5.2 Workflow (just copy me)

The video explains a coding workflow that combines Fable 5.2 for difficult planning and reasoning, Opus 5.5 for faster execution, and Gemini 4 Carbon for independent reviews. It matters because Fable is highly capable but consumes limits quickly, so model selection and task-based routing reduce cost and latency.

VideoDavid Ondrej

More skills = worse agents (do this instead)

The video argues that copying large viral skill repositories can overload an AI agent’s context, waste tokens, and reduce reliability. Instead, keep only a small set of trusted, task-specific skills, create custom skills from recurring workflows, and separate global skills from project-specific ones.

VideoDavid Ondrej

The AI bubble is popping

The video argues that the AI market may be correcting, but the underlying technology is accelerating as frontier models improve and AI adoption expands. For AI builders, the advantage is speed: use leading models across development, security, and operations, and build products that benefit from agents rather than resist them.