30 years of infrastructure. Local AI in the loop. £0 cloud. Watch it work ↓
A local LLM watches the pipeline, diagnoses failures, opens the fix PR.
VS Code · Claude · agents · llama.cpp on a 4090. Zero API spend, zero data leaving.
Proxmox, GitOps, tiered NVMe/HDD storage, tunnels to the edge. Production habits, home prices.