Combining Machine Learning and MLflow with Lakehouse Architecture - Data Brew Episode 5

Explore how Quby leverages machine learning and MLflow with their lakehouse in this Data Brew episode. Learn about Quby's journey from batch to real-time streaming processes for IoT sensor data collection and machine learning. Discover how they extract additional value from their data lake, manage ML models, and implement features like less intrusive monitoring and customer sentiment analysis. Gain insights into the tools, frameworks, and technologies used, including Delta Lake and Spark for distributed training. Understand the implications of dealing with IoT data and how Quby's approach helps them achieve their goal of "outsmarting energy" to make the world more comfortable and sustainable.

Intro
Introduction
How do you save energy
What tools did you start with
What were the implications of dealing with iot data
How has using Delta Lake helped accelerate your process
What type of frameworks are you using
Do you leverage Spark for distributed training
New use cases and features
Less intrusive monitoring
Customer sentiment
Managing ML models
Alerts monitoring
Prior alerts tracking
Performance tracking