Growth & change of careers, organizations & responsibility in data science
summary
Panelists argue concrete practices: one company reports one MS in data science out of 120 on the team, so emphasis is on adaptable skills like building ETL from web logs through front ends, not academic pedigree. Key takeaways include don't hire data scientists without data, avoid over-hiring with no success criteria, keep separate product and data roadmaps, run blameless post-mortems, create a data steward role for provenance and governance, and invest in leadership training (a proposed “leadership carpentry”) and mental‑health safeguards.
AI-written summary · about
chapters (24)
- 0:00Introduction
- 0:20About me
- 0:57The panel
- 4:25The most important lesson youve learned
- 8:07Advice for junior data scientists
- 11:01ruthless prioritization
- 14:32leadership
- 16:24investment in leadership
- 17:41leadership vs work
- 23:54skills atrophy
- 26:50organizations
- 32:13over hiring
- 34:27responsibility in data science
- 39:01good data science teams
- 41:07data ethics
- 41:45data science ethics
- 42:51ownership is critical
- 43:49most effective way to work together
- 44:13being a great partner
- 45:11roadmaps
- 45:47common ways data scientists fail
- 46:22what has happened to that
- 49:09failure in data science
- 50:23Peter Principle
related talks
44mCareers in Data Science
46mPanel Discussion: Business Side of Data Science
42mPanel Discussion: Improving Industry-Academia Collaboration To Close The Data Science Talent Gap
05mWhat I Wish I Knew Before Becoming a Data Scientist
01h 03mHow to think about your career in Data science
55mBuilding The Data Dream Team
21mBeyond the Classroom: Unspoken Realities of a Data Science Career
54mThe secret sauce of data science management