Evaluating Agentic Systems | Roundtable by Rakesh Maram, AT&T [RSVP REQUIRED]

About this session

This event requires an RSVP; to attend, submit form hereHow do we know if an autonomous system is truly capable, rather than optimized for a benchmark? This roundtable will explore why static evaluations are no longer enough for modern AI agents. The discussion will focus on dynamic testing environments that capture reasoning, tool use, failure recovery, and long-horizon planning. I will share practical ways to define success criteria and identify where agents break down before production.