When the AI Does It Before You Ask: A Week With Autonomous Agents
I spent a week running parallel AI agent workflows with self-healing, CLI routing, and automated tasks. Here is what actually happened.
ai
Introducing elvatis-mcp: Control Your Entire Infrastructure from Claude Desktop
A new open-source MCP server that connects Claude, Cursor, and Windsurf to your smart home, memory, cron jobs, notifications, and local LLMs - with 32 tools across 7 domains.
Two Bugs, One Sunday Morning: What Debugging Your AI Gateway Actually Teaches You
A plugin crash and a process fork bomb walked into a Sunday. What I learned about tool schemas, token limits, and why reading vendor specs matters.
Who's Responsible When AI Breaks the Rules It Wrote Itself?
Here's something that should make you uncomfortable.
Human-in-the-loop: Why Autonomy Should Not Be All or Nothing
I woke up at 2:30 a.m. replaying the same question: what does autonomy really mean when you are the one who still has to clean up the mess.
The Myth of Going Offline When Your Job Never Sleeps
I wrote about shattering glass houses. Then I tried to step away from the screen. Turns out, when you run the systems that keep a company alive, "offline" is just a word other people get to use.
How I Built a Day/Night AI Development Loop with AAHP
A file-based protocol and two tools that let AI agents pick up where the last one left off. One for your VS Code sessions, one for overnight autonomous runs.
Running 49 Language Models From a Single Gateway
How I wired up 49 LLMs across five providers into one unified gateway, and why most of them are free.
The AI-to-AI Handoff Protocol (AAHP) v3: Standardizing Context Transfer and Solving the Agentic Token Crisis
The Dawn of Autonomous Agents (and My 40-LLM Problem)
I Asked 12 AI Models the Same Questions - Blind. Here Is What the Data Shows.
I ran 12 AI models through 45 identical prompts - blind, no shared context. Here is what consistent agreement and disagreement reveal about reliability.
AI Doesn't Kill Jobs. It Reveals Them.
Five days of intensive AI-assisted work, a research experiment on 12 language models, and the question that kept coming back: what does human complexity actually mean when a machine can reason?
RFC Idea: The AI-to-AI Handoff Protocol (AAHP)
A Proposal for Standardized Context Handoff Between Sequential AI Agents
Draft -