
Most teams get handed an AI tool and a login. Nobody tells them what happens next.
I'm Michael Brizendine, and what happens next is the part I work on. I run delivery teams on large digital programs, where a careless mistake lands in front of a federal client. Lately a big piece of my job has been integrating AI into those teams in a way people trust enough to use on real work.
What I do all day
I manage people before I manage anything else. I hire them, I coach them, and when a team is not working I am the one who has to fix it. Every result that I am proud of came from a team that was set up well, and I have never seen a tool make up for one that was not.
Getting AI approved turned out to be harder than getting it to work. It took months of patient conversations with a federal client before my teams were allowed to use agentic AI tools at all. When the answer finally came back yes, I wrote the ground rules and the training myself, because a yes with no guardrails behind it tends to get rescinded.
I also like delivery to be a little boring. Work moves through a Kanban board, capacity gets planned before the month starts, and when I give a client a date, I expect to hit it.
Things I learned the hard way
About once a month I write up something from the work that I wish I had known sooner. The first entry came from watching smart people use AI the same way they use a search engine, which wastes most of what it can do.
Your Team Is Using AI Like a Search Engine. Here's How to Fix That.
A little about me
I spent years coaching teams on how to deliver, and now I lead the programs and the people who run them. That path took me through commercial software, financial services, and federal digital work. I live in North Carolina, and if I am not at my desk there is a good chance I have a guitar in my hands.