From simulation to insights: Testing AI agents

​5 March 2026 · 16:00–17:00 CET / 10:00–11:00 EST
Speaker: Dr. Harry Cruz, CTO @ Rhesis AI

​Most AI agents look great in a demo. The real test comes when users start asking follow-up questions, switching context mid-conversation, and pushing the system in directions you never anticipated.

​In this webinar, we'll walk through how to move beyond single-turn evaluation and test your agent the way real users actually interact with it -> across multiple turns, with tool calling, context, and reliability all on the line.

​We'll cover:

• ​Why single-turn testing gives you a false sense of confidence

• ​How synthetic multi-turn conversations expose failure modes you'd otherwise miss

• ​How traces help you understand not just what your agent said, but why it behaved the way it did

• ​A live demo of Penelope, Rhesis's open-source multi-turn testing agent, in action

​Who should attend:

​Whether you're implementing a customer-facing chatbot, an internal AI assistant, or a complex multi-agent system, this session will give you a practical framework for identifying if your agent actually works. Particularly helpful for:

• ​PMs who want to turn user stories and acceptance criteria into test cases

• ​AI engineers wondering how to integrate conversation simulation into their CI/CD pipeline

• ​QA leads & test engineers looking to extend their existing testing practices to cover LLM and agent behavior

​Online webinar. Please register for more details.