How to manage data-related projects and not fail (too often)?
summary
Compares CRISP‑DM and Scrum, argues for Kanban over Scrum for exploratory data work, breaks roles into data analyst/data scientist/machine‑learning‑engineer and shows common gaps, measures effort as 5–20% modeling time with most time on wrangling and deployment, gives cost examples — model training can run tens to $65k, stresses need for labeled data and suggests transfer learning, cloud APIs, and AutoML as low‑cost experiments before full projects.
AI-written summary · about
chapters (10)
- 0:00Introduction
- 1:13Why data projects fail
- 4:06Data vs software processes
- 7:36Roles in data teams
- 11:08Hidden ML costs
- 13:57Automating with models
- 17:09Project costs and data
- 20:44Small data experiments
- 21:49Science and team management
- 24:26Q&A
related talks
43mAgile Strategies for High-Impact Data Science & Machine Learning Projects
31mWhy Most Data Projects Fail and How to Avoid It
39mHelp Me, Help You: Supporting Your Data
54mThe secret sauce of data science management
26mHow to effectively manage AI/ML projects
30mDevOps For Data Teams
16mRedefining Agile for Data Science: A Team First Approach
29mBuilding successful data teams in the realities of 2023