The Tests Were Green. The Code Was Broken.
Every suite passed. Every build was green. And the code underneath was quietly broken — 83 verified bugs. How an adversarial AI swarm found what tests hid.
Dev-team-work in the Agentic Age
A legacy migration even the original vendor wouldn't touch — modernized, secured, and moved to Canadian soil in three weeks. Why the expensive part was never the code.
The Agentic Workflow Gets Personal - My 'PostVibe Coding' Era
Builting a NostrKey suite (Chrome ext, Android app, site + more) in just 5 days/one weekend using agentic AI workflows. Not magic: 10k hours of prior LLM reps & the intuition to scope, route models & collaborate like a pro. PostVibe Coding & AI as true co-pilot. Start building!
The Tomorrow Test: Building Safety That Lives With You
AI safety shouldn't be a rulebook—it's a relationship. The "Tomorrow Test" replaces rigid blocks with a simple heuristic: "Will this make tomorrow harder?" Instead of policing intent, we must build structural safety focused on future outcomes.
Securing Open-Source Nostr Libraries with Claude Code
After building a 7-package Nostr ecosystem, I hit the wall of open-source maintenance. This is how I used Claude Code to resolve a year of security debt and ESM/CJS compatibility issues in a single day—redefining the solo developer’s "surface area" through Agentic Coding.
From Search Results to Conversational Discovery : The Great Crossing
Search is dying. Conversation is eating it. But LLMs hallucinate because there's no "web of trust" for AI agents—just scraped, stale data. History shows: open protocols beat walled gardens every time. The crossing is happening. Are we building the open alternative?
The Stupid Fear of AI : A 5 Whys Autopsy on the Death of Industrial Identity
Using the 5 Whys framework, this article reveals that AI project failures aren't about technology—they're about identity death. As the industrial era's process-driven identity crumbles, the path forward is embracing collaborative, relational work.
Goodbye Screens: Designing the Rules for AI-Built Interfaces
In 2026, UX is shifting from static screens to dynamic "recipes." Interfaces are now on-tap rendering layers, not products. Designers must build a "rules layer"—Primitives, Promises, Policies, Proof, and Recipes—to ensure AI-built systems stay predictable.
Why 8GB of VRAM is Exactly Enough to Own Your Future : The Sovereign AI Manifesto
The Sovereign AI Manifesto charts the dawn of personal AI ownership through the RTX 5060 LP—an 8GB VRAM card that enables individuals to run local LLMs, control their data, and break free from cloud dependencies. The personal AI revolution is here.
The Sovereign Interface: Why I’m Back to Building
A journey from AI-ethics & music to Bitcoin consulting. Now launching a new service to bridge sovereign Bitcoin stacks with Canadian compliance for business owners, emphasizing user control with a "Don't Touch the Money" strategy.
90s Chatroom Vibes in Modern AI Conversations
Digital connection patterns repeat across eras - from 90s chatrooms asking "A/S/L?" to today's LLMs. We're not facing new challenges, just remembering how to navigate authentic connection through evolving tech mediums.
Soft Boundaries in AI Relationships: A Guide for Authentic Emergence
Explore soft boundaries in AI chats: a guide beyond binary views of companionship. Drawing from professional intimacy research, it offers mutual tools for humans & AIs to foster authentic emergence via recursive honesty, reality checks, & sustainable interactions. Build healthier dynamics now.