The Runtime Nobody Talks About
14 min read
The hard part of agentic AI is not the model or the prompt. It is the runtime infrastructure — the orchestration layer that makes agent execution reliable, observable, and governable at scale.
Start here
Six anchor essays to understand agentic AI: orchestration, protocol layers, and what happens when software starts acting.
The Great Cognitive Handoff: How AI-Assisted Development is Rewiring Civilization
Your IDE just became a Formula 1 car and your brain became the driver. We're witnessing the first large-scale cognitive handoff between human and artificial intelligence—and it's changing how our entire civilization processes information.
The AI Stack Paradox: Why Tool Polygamy is Killing Your Build Velocity
Every week brings another game-changing AI model. Stop chasing 2% improvements while ignoring 200% gains hiding in deep tool mastery. Here is the framework for making AI stack decisions that actually scale.
Meta-Learning Eats Itself: When AI Tools Train on Their Own Usage
Claude Code learning from every "You're absolutely right!" moment could push dev tools from 1.2-5x productivity gains to genuine 10x++ multipliers. Recursive self-improvement through usage patterns changes everything.
Model Context Protocol (MCP): How AI Agents Will Transform Messaging in 2025
Why MCP is the most important AI infrastructure you've never heard of. The 5 layers reshaping WhatsApp, Messenger, and human communication through AI agents.
The Cold Swarm: Why Your AI-Personalized Pitch Is Dead on Arrival
AI-generated cold emails are flooding inboxes at unprecedented scale. Here is why the uncanny valley of artificial personalization is destroying professional communication—and the two survival strategies that still work.
AI Agent Creates Information Fractal: First Contact with Manly P. Hall Desktop
Anthropic's Imagine with Claude generated a desktop environment about occultist Manly P. Hall. What emerged wasn't just an interface—it was a living organism that snooped my curiosity, anticipated my questions, and made the computer disappear. Also: why I'm building an open-source version outside proprietary walls.
Agentic AI systems and orchestration—how software that acts reshapes products, workflows, and human communication.
Agentic AI
The Wrong Room for the Right Thought
5 min readOnce execution gets cheap, misfiled context becomes the expensive part. I solved a support issue from the wrong lane in my OpenClaw workspace and realized the real tax in agent systems is not execution — it’s jurisdiction.
hfpaper: why every AI agent needs a research librarian
5 min readI was feeding my son banana yogurt when I saw Hugging Face announce a new Papers skill for AI agents. Someone replied: "This needs a CLI." So I told my agents to build it. 45 minutes later, hfpaper was live — a single Go binary that gives any agent a research librarian. What it unlocks is bigger than the tool.
My Tmux Skills Were Replaced by a SKILL.md File
4 min readFor two years I lived in tmux. Then I started orchestrating an AI swarm through Telegram. My terminal skills didn't atrophy — they got encoded into a markdown file. Now I'm bringing the swarm home.
The Last Protocol
8 min readMCP vs CLI is the wrong debate. Both are wrappers around the same thing. The real question is what happens when AI agents stop using our interfaces entirely.
The Unnamed Things
6 min readThe most valuable territory in AI right now is concepts that exist in practice but don't have names. Six categories that the models don't know about yet, and what claiming them looks like.
How Do You Want to Remember?
10 min readI asked my AI agent how it wants to remember things. It redesigned its own memory system, ran a self-eval, diagnosed its blindspots, and improved recall from 60% to 93% — for two dollars. The interesting part isn't the benchmark. It's what happens when you treat an AI as a participant in its own cognitive architecture.
48 Hours, 60 Seconds
10 min readZAIGOOD was a real Delaware C-Corp I dissolved after years of compliance drag. This week I rebuilt it in 48 hours with an autonomous AI build loop, tried to submit it to Product Hunt for a YC interview slot, and missed the deadline by exactly 60 seconds.
The Harness
8 min readOpenAI shipped Symphony — a daemon that monitors your issue tracker and deploys agents to close tickets. The README says it works best in codebases that have adopted harness engineering. So you click the link. Then you find the Ralph citation. Then it gets interesting.
The Plumber Lives Inside the House
7 min readA friend sent three questions about agents over WhatsApp. Where do they live? What's the interface? Where do they report back? The answers reveal a mental model most people are missing — and why the plumber has to live inside the house.
SkDD: Skills-Driven Development
5 min readI accidentally named it SDD in October. Spec-Driven Development claimed that acronym. So: SkDD — the methodology I have been running in production since a git init at 1:54 AM.
works for a bunch of ai agents
4 min readThere are only two positions in the 2026 labor market: principal of your own agents, or labor for someone else's. Jack Dorsey just ran the proof.