Galileo Video: Architecting Reliable Agentic AI
This is not about my own work — it is a talk I recommend because it is directly relevant to the broader question of building reliable agentic systems.
Giovanna Carofiglio gave a talk for Galileo in July 2025 on reliable agentic AI: what it takes to build agent systems that remain inspectable and trustworthy under real operational conditions.
The talk covers engineering concerns — observability, failure modes, trust boundaries — that matter to anyone deploying agents in production, regardless of the specific domain.