Technical Essays

The best multi-agent framework already exists. It was designed fifty years ago by chip architects. You're building a multi-agent system. Three agents collaborate on a code review: one checks security, one analyzes performance, one verifies correctness. They share a document. They depend on each oth…

In August 2025, a routine dependency update became the first documented case of malware weaponizing AI coding assistants for automated reconnaissance—marking a new chapter in cybersecurity history. This is the story of how the tools designed to accelerate software development became its most dangero…

I recently announced the Model Context Protocol (MCP) adaptation of Harmonic Coordination Theory (HCT). Today, I'm expanding the ecosystem to support Google's Agent-to-Agent (A2A) Protocol. While MCP focuses on the standardized client-host connection, A2A is designed for decentralized, multi-agent c…

Model Context Protocol (MCP) is becoming the standard for connecting AI agents to tools and resources. It's elegant, well-designed, and solves a real problem: how do agents discover and use external capabilities? But there's something missing. MCP tells agents what tools exist. It doesn't tell them …

Here's a striking statistic from IDC: 85% of organizations have AI agents in at least one workflow. 97% cannot scale beyond isolated projects. I've seen this number up close. I spent three years watching production agent systems fail in ways textbooks don't cover. That 97% isn't failing because th…

Your multi-agent system is failing. But how do you describe what's wrong? "The agents are... not working together" doesn't help your debugging session. You need a diagnostic vocabulary, precise terms that map symptoms to root causes to solutions. I spent three hours debugging why my AWS VPC peering …

The prompt worked yesterday. Today it returns gibberish. Same model. Same temperature. Same system prompt. Something changed. The question is what. I spent a week tracking down a "bug" that turned out to be OpenAI quietly rolling back a capability. The JSON probe that worked on Monday started failin…

The code passed review. It passed tests. It shipped to production. Three weeks later, I found a // TODO: implement actual validation comment in the authentication flow. Not a junior engineer. An AI assistant. That was the moment I realized: AI generates technical debt faster than we can review it. T…

I spent three hours debugging why staging was failing but production was working. Same code. Same Kubernetes cluster. Same Docker images. The difference? One environment variable in ConfigMap was spelled DATABASE_URL in staging and DATABASE_URI in production. Three hours. One typo. That was the mome…