Session: Practical Lessons from Building and Scaling AI Systems
This talk shares real-world lessons from building and scaling AI systems in large consumer products. Drawing on experience from recommendation and generative AI use cases, it covers how to move from promising models to reliable production systems, balance offline metrics with online impact, and navigate challenges around data quality, experimentation, and global user behavior. The focus is on pragmatic takeaways that teams can apply when deploying AI at scale.
Bio
Ishita is a Senior Machine Learning Engineer at Netflix on the Core Recommendations team, where she designs and scales AI systems that power member experiences and large-scale personalization worldwide. Previously, she worked as a Senior Machine Learning Engineer at Adobe on the Globalization team, leading the development and deployment of AI and generative AI capabilities across products. Her work centers on high-impact recommender systems, generative AI, and robust, scalable production ML