About

Think agentic. is a blog about building AI agents: software that doesn’t just answer questions, but takes action, follows a process and gets real work done.

I’m Dan Finan. I’m a Microsoft automation consultant, and I spend my days designing and building agents, automations and the governance that keeps them safe. Before that I was a teacher, which probably explains why I can’t learn something without wanting to explain it.

Why it exists

Agentic AI is moving fast, and most of what’s written about it is either hype or reference documentation. There’s very little in between: honest, practical accounts of what it’s like to build these things for real, including the parts that don’t work first time.

This blog is my attempt to fill that gap. Every post starts with something I’ve built, broken, fixed or wondered about.

Who it’s for

Builders of every level.

If you’ve never built an agent, you’re especially welcome. You don’t need to be a developer to start. Today’s tools put real, useful agents within reach of anyone curious enough to try, and I hope something here makes you want to.

If you’re further along, expect patterns, trade-offs and lessons you can take straight into your own work.

What you’ll find

  • Walkthroughs of real builds, from first idea to working solution
  • Decisions and trade-offs: low-code or pro-code, and when an agent is the wrong answer
  • Governance, security and the unglamorous work that keeps agents running
  • Notes from the field: what I’m learning, as I learn it

Why Microsoft

Most of my work happens in the Microsoft stack: Copilot Studio, Power Platform, Azure AI Foundry, and Python when the low-code tools run out. That’s where most of the writing lives. The thinking behind it applies wherever you build.

Written by a person

The articles here are written by me, not generated by AI. That’s deliberate. Writing is how I learn, and explaining something properly is the best test of whether I actually understand it. So this blog is as much a learning tool for me as it is for you. I’ll get things wrong sometimes, and I’ll say so when I do.

I use AI in my work every day, and plenty of posts will be about exactly that. But the words are mine.