As data collection and analysis continue to expand, the need for effective and transparent anonymization becomes increasingly critical. Yet the growing diversity of anonymization techniques makes it difficult fo r developers to ev aluate th eir guarantees and choose an appropriate method at design time. We present AnonymEx, an interactive platform that enables developers to empirically compare anonymization techniques using a unified metric: their susceptibility to re-identification attacks. The platform integrates (i) a literature-grounded knowledge graph linking anonymization techniques to documented re-identification attacks, (ii) executable, containerized implementations of both techniques and attacks, and (iii) an assistant that supports exploratory learning and helps users identify candidate techniques based on their requirements. During the demonstration, attendees explore the anonymization landscape, test techniques using provided datasets, run the associated attacks, and assess the assistant’s suggestions within the Anonymization-by-Design workflow. AnonymEx thus provides a practical, transparent, and reproducible sandbox for design-time decision-making and crosscategory empirical evaluation of anonymization techniques.