What I have my head in this quarter. This page changes often, and that is the point.
Taking AI agents into real operation, not into a demo: the limits of what an agent may decide alone, where the person sits, and how each decision is audited.
And an old obsession I cannot shake: that technology should leave capability behind. Teams that know how to decide when I am not there.
Cases of applied AI in regulated industries, mostly banking and healthcare, where being wrong is expensive and the solutions are therefore more honest.
Architecture of agent-based systems: orchestration, limits of autonomy and traceability of decisions.
Open the chat and tell me. I am interested in comparing notes.
This is a "Now" page: a personal-site convention for telling what someone is up to today, not only what they did before.
An open conversation to share knowledge. Messages reach my console and I reply from there.
Pick the day and time that work for you. Thirty minutes, no sales pitch.
For a video call, just ask for one here and I'll send you the session link.
