Explore cloud directions, MLOps, and production data science in this 43-minute video featuring Joseph M. Hellerstein, Professor of Computer Science at UC Berkeley. Delve into key topics including model development, training pipelines, inference, and the importance of data in the ML lifecycle. Examine three critical challenges in MLOps and gain insights into cloud programming and serverless computing. Learn about the RISE of Aqueduct and its impact on tech transfer. Discover practical takeaways for implementing MLOps best practices and leveraging cloud-based solutions to scale data science models while ensuring reliability, maintainability, and scalability in your organization.
Cloud Directions, MLOps and Production Data Science
Overview
Syllabus
- Introduction
- Outline
- Model Development
- Training Pipelines
- Inference
- What’s the Data?
- Data is Life! The Virtuous Cycle
- Next: 3 Key Challenges in MLOps
- Takeawys: The ML Lifecycle
- Cloud Programming & Serverless Computing
- The Big Question
- A Taste of Tech Transfer: the RISE of Aqueduct
- Aqueduct Uswer Interviews
- Takeaways
Taught by
Open Data Science
Related Courses
-
MLOps Platforms: Amazon SageMaker and Azure ML
-
Introduction to MLOps: What, Why, and How
-
Combatting Illegal Fishing with Machine Learning and Azure
-
Unpacking Three Types of Feature Stores - MLOps Podcast
-
The State of Production MLOps in the Cloud Native Ecosystem
-
Building Real-Time ML Pipelines the Easy Way
