Building robust data lakes with Apache Flink often leads to a crucial decision: selecting the right open table format from options like Apache Hudi, Iceberg, and Paimon. This session cuts through the common confusion, providing a head-to-head comparison of these seemingly similar technologies. Attendees will learn the three key distinctions vital for making an informed choice: how each format organizes data under the hood, their respective optimizations for various update and query patterns, and the nuances of their integration with Flink. The talk equips engineers with the practical understanding needed to confidently select the best-suited format and build an optimized Flink-powered data lake architecture.