Agents in the Room
Production notes on agentic AI, for people who decide.
Written from live projects: what held up, what broke, and what it changes for the decision on your desk.
What's in each email
One idea to act on
A thesis from current production work, with the reasoning behind it, written so you can bring it to your next management meeting.
One lesson from the field
A situation from a real project, anonymised: what we set up, where it broke, what I do differently the next time.
One signal from the stack
A change at Google, at Anthropic or in the protocols (MCP, A2A) that touches a live project, and whether it should change one of your decisions.
Each issue ends with the three points I would raise if I sat in your board meeting.
Who it's for
Founders, CEOs and CTOs who have to decide on AI: what to build, who should lead it, what deserves attention this quarter. Short enough to read between two meetings, specific enough to disagree with.
Who's writing
I'm Matteo Gazzurelli, fractional CTO and agentic AI consultant. I run MaGa Srl from Brescia and decide, with companies that have repeatable processes, where an AI agent earns its place; the newsletter is what I learn on the way.
Schedule and privacy
Issues go out when there is something worth reading. Every issue includes an unsubscribe link. See the privacy policy for how your data is handled.