Leading Data Science Teams - A Framework to Help Guide Data Science Project Managers - Jeffrey Saltz

Explore a comprehensive framework for leading data science teams effectively in this 46-minute video presentation. Learn how to address key challenges faced by data science managers and senior leaders, including project execution, data and algorithm misuse prevention, and result validation. Dive into essential aspects of the framework, such as forming data science teams, establishing processes for developing analytical solutions, and implementing risk management strategies. Gain insights into team roles, project management methodologies like Scrum, CRISP-DM, and Kanban, and ethical considerations in data science. Discover techniques for ensuring data quality, managing analytics workflows, and maintaining production robustness. Enhance your ability to guide data science projects successfully and efficiently with practical examples and key questions to consider throughout the development process.

Intro
INTRODUCTION
OUTLINE
THINKING ABOUT PROJECT RISKS Some possible Issues
MANAGING PROJECT RISKS Example questions to ask
KEY AREAS OF FOCUS: 1 Team Roles - The composition of team
Data Scientist uses visualizations and machine learning to aid in the understanding data. Has an overview of the end-to-end process
EXAMPLE TEAM ROLES
TEAM PROCESS The process the team uses
TEAM PROCESS: SCRUM
TEAM PROCESS: CRISP-DM
TEAM PROCESS: KANBAN
ETHICS
EXAMPLE ETHICAL QUESTIONS
HOW TO ENSURE DATA QUALITY Some key questions: • Did cleaning introduce errors?
ANALYTICS WORKFLOW EXAMPLE
MODEL MANAGEMENT
PRODUCTION ROBUSTNESS
EXAMPLE MODEL DASHBOARD
2 QUICK POLLS - PRIORITIZATION & ROLE