What if process intelligence could evolve in real time — not just describe how humans work, but how agents work too?

In this technical deep dive, Andrew Wolfe (Co-CEO, Bloomfilter) unveils Agentic Process Miner, a system built during Celonis’ hackathon that uses an MCP (Model Context Protocol) server to mine the actions of AI agents, analyze them through Celonis process intelligence, and deliver insights directly back to the agents themselves.

Attendees will see how the team built a feedback loop between agent execution and process mining, enabling continuous optimization and transparent monitoring of autonomous workflows. Expect code-level details, architecture patterns, and a glimpse into the future of self-improving, data-aware AI ecosystems.