Transform Your Machine Learning Pipelines with Apache Hudi

Discover how to revolutionize machine learning pipelines integrated with data lakes in this 25-minute conference talk by Nadine Farah from Onehouse. Learn about the challenges of maintaining fresh, accurate, and near real-time data for ML models in traditional data lakes. Explore how Apache Hudi addresses these issues with features like upserts, incremental processing, and near real-time access. Gain insights into building efficient ML pipelines using Hudi's capabilities, including time-travel querying and incremental data pulls. Understand how to overcome data latency, implement incremental updates, and ensure timely data availability for ML models. By the end of this talk, acquire knowledge on transforming your ML pipelines to harness the full potential of data lakes using Apache Hudi.