18/08/2026
π Evaluating AI Agents: From Outputs to Real-World Performance
How do we know if an AI agent actually works? π€
Join us for an insightful workshop with Sabaa Melliti, where weβll explore how to evaluate AI agents beyond their final answers β from BLEU, ROUGE & BERTScore to RAGAS, Task Success, Tool-Use Accuracy, Trajectory Evaluation, Robustness & Efficiency. π
π
Tuesday, August 18
π 23:00
π» Google Meet
Discover how to measure what an AI agent does, not just what it says. π€β¨
π Donβt miss it β see you online!