9 posts
How we built the Phase 1 foundation for Canonical v2: brokered writes via a staging log, strict validation, quarantine, a ledger index, and namespace guards to prevent silent failure.
When AI agents write files autonomously, you need something watching them. Here's how I built ForgedWatchdog — a configurable Python watchdog with layer-based detection, classification, and auto-pause.
How I built an automated podcast production pipeline — AI content generation, audio synthesis, episode formatting, and publish-ready output from a structured brief. A case study in end-to-end media automation.
How I built a real-time multi-agent monitoring dashboard in PyQt6 — carrier scene visualisation, live status feeds, and autonomous system oversight from a single control panel.
How I built a local AI assistant from scratch — LLAMA integration, custom PyQt6 interface, and a local-first architecture that keeps data on your machine. A case study in practical AI deployment.
How I designed and built a Discord-integrated org management system for Star Citizen — from database architecture to live web dashboard. A case study in full-stack development and practical SQLite design.
I run a persistent AI automation system on a home server — multiple agents, scheduled work, 24/7 operations. Here's what the reality looks like: the wins, the failures, and what I've learned.
I built a multi-agent AI system that handles task management, code deployment, content generation, and monitoring — running 24/7 on a home server. Here's what that looks like in practice.
How I turned 9 months of org economy data into a trading intelligence system — with real numbers from the SoakHauler database.