I run a homelab focused on local AI experimentation, infrastructure automation, and the occasional hardware project. This site documents what's running, what broke, and what's next.
Building a dedicated GPU server for local LLM inference and fine-tuning. Started with a single RTX 3090, planning to add a second card once cooling is sorted. Runs Ollama, vLLM, and custom agent workloads in Docker containers.
Migrated from a bare-metal Ubuntu server to Unraid for better storage management and container orchestration. The entire homelab now runs as Docker containers managed through docker-compose files versioned in Git.