During November 2024, Andrey Gubichev developed a customizable query optimization hints framework for the xupefei/spark repository, focusing on aggregate, sort, and window operators. He designed an abstract hint mechanism for aggregates and introduced an extension API that allows custom hint resolution rules, enabling greater flexibility in guiding the optimizer and execution planning. By enabling hints on Sort and Window nodes, Andrey improved the configurability and extensibility of Spark’s optimizer, laying groundwork for future performance enhancements. His work leveraged Scala, Spark, and SQL, demonstrating depth in both architectural design and practical implementation, with a clear focus on extensible optimization features.