Local-AI Automation Lab

I lead high-performance technical teams and run the infrastructure to prove it — self-hosted, AI-assisted, no hype, just things that work.

This is the proof layer for my content: every post I publish points at a folder here with the actual artefacts — n8n workflows, prompts, Dockerfiles, benchmark notes, and honest failure logs. Built on my own hardware (TrueNAS + RTX 4060, Mac M5, two OptiPlexes), orchestrated by n8n, drafted by local Ollama models.

Structure

• posts/ — one folder per published post: README, workflow JSON, prompts, configs
• learning-log/ — daily learning notes (leadership + platform engineering)
• templates/ — reusable n8n workflows and pipeline stages, promoted from posts/
• docker/ — container definitions for the GPU render pipeline
• docs/ — architecture, licences, decisions

The pipeline that publishes this repo

Sources (course notes, saved links) → local LLM drafting → quality gate → confidentiality scrubber → human approval → multi-platform publish + this repo. Engagement is never automated; every post is approved by me.

AI-assisted content, my real face and words. Everything here runs locally unless explicitly noted.