Introducing Kubeflow for Machine Learning and AI on Kubernetes

Explore the world of Kubeflow in this informative conference talk. Discover how this open-source project simplifies the process of moving machine learning models from development to production. Learn about the challenges faced by data scientists and engineers in deploying models, and how Kubeflow addresses these issues. Gain insights into the components of Kubeflow, including model training, persistence, deployment, and serving. Understand the importance of scalability in training and model deployment, and how Kubeflow integrates with the broader ecosystem. Delve into topics such as pipeline execution, scheduling, and metadata management. Examine the potential downsides of Kubeflow and get a glimpse of cross-cloud workflows. Enhance your understanding of how Kubeflow can streamline the journey from model creation to production deployment in the ever-evolving landscape of data science, machine learning, and artificial intelligence.

Introducing Kubeflow
Trevor Grant
What is Kubernetes?
What is Machine Learning/AI.
ML Tools are making DS suck less and less every year
Problems of Data Scientists and ML Idiots
What is Kubeflow?
Components Buffet
The many kinds of models you can train
But don't forget about data prep friends!
Model persistence/deployment/c
Model Serving
How does this fit in to the Ecosystem
Running that pipeline
Execute Pipelines on a Schedule
Better Pipelines with Python Code
Meta Data!
Downsides to kubellow
Very DIY KF Cross Cloud Workshop
Questions? Half-baked Demo?