Abstract: The largest challenge in ML today is effectively deploying reliable and efficient models into production, with experts quoting that as many as 90% of model created never make it to production. MLOps streamlines the process of taking machine learning models to production, and then maintaining and monitoring them. With new MLOps micro-venders popping up every day, what toolset should you be using?
In this session, we will consider how Delta Lake can power feature stores, model registries and model serving, with Databricks and MLFlow providing the environment. Delta streamlines the processes of capturing training datasets across experimentations and is the foundation for a flexible, agile lake.
Covering Delta Lake, Databricks Feature Store, MLFlow and real-time Model Serving, this workshop is suitable for Data Scientists and Machine Learning Engineers of all levels.
Bio: Alex is a data scientist at Advancing Analytics with a love for all things machine learning and MLOps. He has worked as a machine learning engineer for four years in fields ranging from wearable devices to agritech. Outside of the world of data science he is an avid fan of board games and TTRPGs, running a number of D&D campaigns and a small Lincoln based D&D community.