Let's Get Aggregated: Custom UDAFs in Spark

Tired of the same old avg() and sum() in Spark? Wish your data could tell more interesting stories? Then this talk is for you! We’ll dive into the world of User-Defined Aggregation Functions (UDAFs) in Apache Spark — your secret weapon for solving those “just-one-more-metric” moments that built-ins can’t handle. You’ll learn how to build your own aggregations using Spark’s type-safe Aggregator API, giving you full control, better performance, and compiler-approved confidence. We’ll show you how to craft custom metrics that actually reflect your business logic. Whether you’re wrangling massive datasets or just want to impress your coworkers with some next-level Spark wizardry, join us and level up your data game — one custom aggregation at a time!