Learn how to easily set up and run your own coding assistant model in the cloud using Hetzner servers, NVIDIA GPU drivers, CUDA, and Ollama. The tutorial covers essential steps, from initial server setup with SSH keys to installing necessary drivers and software.
Additionally, discover how to enhance security and control by integrating TailScale VPN, managing access with UFW firewall, and exposing models to the internet with TailScale Funnel for seamless access through coding assistants like Cursor.
Ideal for developers seeking to manage their own coding infrastructure with full observability and security.
00:00 Introduction to hosting your own coding assistant
00:55 Setting up the server
01:47 Installing drivers and CUDA
08:35 Configuring Ollama for coding assistance
12:38 Securing the server with VPN and Tailscale
20:02 Integrating with Cursor and TailScale Funnel
24:15 Conclusion and next steps