From AI agents to agentic applications — at APMS 2026
// Speaking
AI agents are getting better fast. The bigger question is how to turn them into something companies can actually use in daily operations.
The application, not the agent
In operational environments, AI needs to work together with business data, deterministic logic, tools, user interfaces and human decision-making. None of that is the model — and all of it decides whether the result is usable.
That is what the paper examines: the architecture around the agent, worked through a case in inventory management.
Where the value actually comes from
This is one of the key shifts in enterprise AI. The real value will not come from the LLM alone, but from how AI is embedded into the systems and processes where work actually happens.
Looking forward to the discussions in Bergamo.
One hour is enough.
Give us an hour on one of your processes. The next day you see a working prototype of it — your workflow, your screens, your logo.