Evaluation of AI systems is becoming increasingly complex as the use cases for artificial intelligence are expanding. The current industry standard, LLM-as-a-Judge, may not be sufficient to meet the demands of this new landscape. New approaches, such as Evaluation Agents, are being explored to improve the effectiveness and usefulness of AI evaluations. Research by Galileo's team has uncovered some interesting findings in this area, which will be discussed during an upcoming webinar.