Q&A with Keynote Katrina Riehl
summary
Recommends concrete steps for healthier projects: require affiliation with NumFOCUS or at least a code of conduct, add a first-contribution guide and follow-up outreach, and run timed bake-offs so teams can compare libraries. Cites a typical job tenure of about 2.5 years as context for retention problems, stresses governance and PR review to avoid proprietary leakage, flags bias in machine learning and adversarial inputs as the highest practical risks, and endorses player‑coach leadership when teams are blocked.
AI-written summary · about
chapters (9)
- 0:00Introduction
- 0:49Diversity in open source
- 5:52Retaining women in tech
- 10:27Proprietary code risks
- 15:45Metrics and communication mistakes
- 21:37Learning versus problem-solving
- 27:19Tool disagreements
- 30:32Ethical ML concerns
- 34:28Data science career paths
related talks
29mQ&A with Keynote – Marit Hansen
44mKeynote: The neglected art of a relevant benchmark and why you need one
34mKeynote: Building for AI at Scale: Infra, Integrity, & Innovation
01h 06mKeynotes 09/18/2021
30mConsumer To Contributor: Lessons Learned
04mThree Questions With...
01mCall for Proposals & Keynote
40mSocially responsible AI: Practical tools for robust design