Blog
Practical insights on building and running AI agent systems in production. No hype, just what works.
What I Shipped in mnemonic Since Launch: A Widget, Honest Time Tracking, and Codex Support
When I launched mnemonic it was an invisible memory daemon. Two months later it has a face — a macOS menu-bar widget that shows worked time, honest per-project hours, and a readable daily journal — plus a security audit and shared memory across Claude Code and Codex.
How We Rescued a Hacked WordPress Site in 3 Hours
What started as a routine SEO audit turned into a forensic recovery — 16 critical vulnerabilities, 5 PHP backdoors, a black-hat SEO agency campaign, and an unpatched Elementor Pro CVE. Here's how we cleaned it in 3 hours.
I Built a Background Memory Daemon for AI Coding Agents
AI coding agents forget everything between sessions. So I built mnemonic — a Rust daemon that captures decisions, corrections, and changes automatically and keeps context persistent across sessions.
How I Rebuilt My AI Dev Stack After Anthropic Killed OAuth
When Anthropic disabled OAuth for third-party tools in January 2026, my multi-agent workflow broke overnight. Here's the dual-stack system I built to never depend on one vendor again.
How I Run 5 AI Agents in Production for $4/Day
A breakdown of my multi-agent system architecture, cost optimization strategies, and the protocols that keep everything running 24/7 without burning through API credits.
AI Agent vs Chatbot: What Business Owners Need to Know
ChatGPT is not an AI agent. Here's the difference, why it matters for your business, and how to decide which one you actually need.
How to Automate Lead Qualification with AI Agents
Step-by-step guide to building an AI agent that qualifies leads 24/7 — scores prospects, sends follow-ups, and books calls while you sleep.