DevOps for Data Scientists

Introduction

• Welcome
• Target audience

1. Data Science Development Practices

• Data science and software engineering
• Collecting and munging data
• Experimenting with data, features, and algorithms
• Testing and validating models

2. Data Science Models to Production

• Version control for data science models
• Predictive Model Markup Language
• Deploying models with automation tools

3. Deployment Practices

• Deploying to staging environment
• Canary deployments
• Securing the data science models in production
• Monitoring models in production

4. Data Science Models in Containers

• Introduction to Docker
• Creating a Dockerfile for data science models
• Data science Docker image repository

Conclusion

• Overview of DevOps best practices for data science