Building ML solutions from scratch is challenging because of a variety of reasons: the long development cycles of writing low level machine learning code and the fast pace of state-of-the-art ML methods to name a few. On the other hand, solutions that automate the ML model development process are often opaque and hard to iterate on, resulting in users churning out. In this talk I’ll cover declarative ML systems, and how they address key issues that help shorten the time taken to bring ML models to production.
Founding Engineer @Predibase
Shreya is a Sr. ML Engineer and Domain Lead for ML Infrastructure at Predibase. Her work involves building scalable solutions for ML training and inference that improve the stability, robustness and effectiveness of ML model training. Previously, she'd worked on using state of the art ML models to solve problems in autonomous systems.