I Built AI Agents for My Actual Life. Let's See If They Work.

You've seen the posts. Someone has automated their entire business, their morning routine, their inbox, their life. Agents everywhere. Frictionless. Revolutionary.

I'm not selling that.

I spend a lot of time thinking about why analytics and AI investments fail or succeed at moving outcomes — in organizations, on teams, at the executive level. The diagnosis is almost never the technology. It's the distance between the signal and the actions driving outcomes - and the people in charge of each.

So I got curious whether the same problem exists at the personal level. And I started building.

Over the last several months I've been putting together a small set of AI agents around my actual decisions — a job search, a home energy project, some civic work, the occasional cocktail. The question I'm testing: does encoding your decision making framework (values, priorities, risks) into a system actually produce better decisions? Or does it just produce more confident-sounding noise?

This session is the experiment, run live with your inputs.

You throw me a real decision. I run it through the agents. We watch what happens — where it's genuinely useful, where it confidently gets it wrong, where the human judgment layer still does all the real work.

Then we talk about what that means. For how you work. For how your team works. For what "decision architecture" actually looks like when it's one person with a laptop or thousands of people in an enterprise with a data platform.

No conclusions yet. That's kind of the point.

Come ready to participate. Bring a real decision if you have one.

• All levels
• AI/ML
• Product
• Soft Skills