DevOps for Data Science

Explore DevOps best practices for Data Science in this 59-minute conference talk. Learn how to apply modern software development techniques to data science projects, addressing often-overlooked aspects such as documentation, data and model refresh, testing, performance monitoring, and security compliance. Discover the tools and practices that enable continuous value addition in data science efforts, bridging the gap between agile software development and predictive model creation. Gain insights into DevOps metrics, processes, and the data science lifecycle, while exploring topics like source control, infrastructure as code, automation, continuous integration, testing, and delivery. Understand how to implement pipelines, shift left, and leverage tools like Jupiter Notebooks to enhance your data science workflow.

Introduction
Who is Damian Brady
Data Science and DevOps
What is DevOps
DevOps Metrics
DevOps Process
Data Science Definition
Data Science Cycle
Data Science Processes
Data Science Life Cycles
Languages and Tools
Operationalization
Source Control
Refactor
Infrastructure as Code
Automation
Continuous Integration
Continuous Testing
Continuous Delivery
Summary
Pipelines
Shift Left
Ignite
Jupiter Notebooks
Data Breach Notebooks
Data Breach Starter Mix
Data Types