featured article
Claude Agent SDK Credit: The Honest Read for IT Leaders
Latest articles
AI Security Triage: The Gap That Is Getting You Breached
June 1, 2026
More than half of organizations with AI security controls still got breached. Here is the real 2026 data on AI security triage and how to close the gap.
AI Agent Instructions Are Only As Good As Your Last Voice Note
May 26, 2026
AI agent instructions can quietly break from one garbled voice note and a missing Yes in their instructions. Here is how I traced April's missing calendar blocks to a single overbroad rule.
What Fox Studios Taught Me About Internal IT Communications
May 18, 2026
Fox Studios taught me a lesson about internal IT communications that I still run today, now scaled with AI minions and three cats with strong opinions.
OAuth Device Code Phishing: The Attack Your M365 Tenant Probably Isn’t Watching For
May 13, 2026
OAuth device code phishing targets users without stealing passwords. Attackers create valid codes, convincing users to authenticate them unknowingly. This method bypasses traditional phishing defenses, exploiting legitimate flows. Detection and prevention involve monitoring sign-in logs and implementing Conditional Access policies.
Why Copilot Voice Chat Won My Commute This Morning
May 12, 2026
My AI minions just learned how to ride shotgun. One of them, anyway, and ... Read more
How I Used AI to Build My Foster Care Tracking System (And How You Can Too)
April 30, 2026
Using Claude Code to build an Airtable base is like using a chainsaw to cut a sandwich. Wrong tool, wrong day. Most IT leaders I talk to have heard of Claude Code (because developers will not stop talking about it) and have not heard of Claude Cowork. That is a problem because Cowork is the one most of you actually want. New post breaks down when to use which, the three-step workflow that gets you a working tool in minutes instead of hours, and a real example I just built.
Meet A.P.R.I.L.: My AI Chief of Staff Agent Who Runs the Operational Layer I Used to Run Myself
April 29, 2026
Here's a strong excerpt for promotional use, around 150 words: There is a version of the inbox problem that every productivity app promises to fix. Superhuman, Fyxer, Microsoft 365 Copilot. They all offer some flavor of prioritization, summarization, or smart drafting. I have used most of them. They help at the email level. They do not help at the operational level. The inbox is not the problem. The inbox is the symptom. The real problem is that information about any given situation is distributed across four or five places simultaneously. A client emails me. There is a ConnectWise ticket behind it. There are Airtable tasks tied to that project. There may be a Teams thread where someone mentioned me and I missed it. By the time I go to respond to the email, I am reconstructing context that should have been handed to me already. That reconstruction is where time disappears, and more importantly, it is where details drop.
Mastering AI Prompts: Transformative Habits for Success
April 28, 2026
Better AI outputs start with better habits, not better models. Here are the prompt habits to build before you automate anything.
AI Is Reflecting You Back. Are You Paying Attention?
April 25, 2026
Using AI as an extension of yourself is powerful, but prompt mirroring and the Believability Effect are real risks. Here's how IT leaders stay grounded.
Receive the latest articles in your inbox
[insert e-mail subscription form]