Back to Projects
Local LLM & Homelab Experiments
Private Proxmox-based environment used to explore self-hosted services, local LLM inference with llama.cpp, automation workflows and developer tooling.
Local LLM Inference
Docker
Proxmox
llama.cpp
n8n
Windmill
Developer Tooling
The Problem
I wanted a controlled environment for learning how self-hosted services, local models and automation fit together.
The Approach
Built a private Proxmox-based homelab and used llama.cpp, n8n, Windmill and containerised services for repeatable experiments.
The Results
A living sandbox for local LLM inference, self-hosting, automation workflows and developer tooling experiments.