Live builds in Claude, ChatGPT, and Microsoft Copilot
Most product managers have gotten good at prompting. They know which tool to open, can get a usable first draft in minutes, and have a few tricks to share with their teams.
But the question has changed. It is no longer whether product teams use AI. It is whether AI is making the work better.
That requires moving beyond one-off prompts. With an agent, you can hand over a recurring piece of work, set clear expectations, and review what comes back. Your judgment stays central. What goes away is the need to rebuild the same process every time.
In this session, Dan Corbin builds agents live in Claude, ChatGPT, and Microsoft Copilot, including Copilot’s new agent mode. Each demo tackles a real product management workflow, so you can see what changes when you delegate the work instead of prompting for another output.
You’ll leave with:
- A framework for deciding which recurring tasks are ready to delegate and where human judgment still matters
- Live examples of agents built for real product workflows, not generic productivity demos
- A short list of useful first agents you can build this week
- A practical way to turn recurring work into playbooks an agent can run
Follow along in whichever tool you already use.
