Building Multilingual Data Science Teams
summary
Argues that R and Python coexist comfortably and that a production multilingual team is practical and valuable, supported by concrete tooling and practices: use renv and venv or Posit Workbench for reproducible OS/R/Python/package versions, pick syntax‑aligned libraries like polars and dplyr, ggplot2/plotnine and gt/great_tables, and use DuckDB or Arrow for language‑agnostic SQL speed. Enforces docs (Roxygen/SciPy docstrings) and Git workflow with issues containing an overview, a reprex, and discussion; the setup runs reliably in consulting projects.
AI-written summary · about
chapters (7)
- 0:00Multilingual DS teams intro
- 1:53Empathy in teamwork
- 2:38Environment management
- 5:04Package choices and syntax
- 9:50Documentation and guides
- 16:25Team collaboration tips
- 16:45Q&A on multilingual teams
related talks
31mProductionalizing a Data Science Team
23mData science as a team sport
20mMastering the Art of Adopting R and Python: Innovative Strategies for Effective Change Management
22mBuild Smarter, Don't Just Build
26mHow to Succeed with Python Across the Enterprise
30mBuilding the best data team is not rocket science
37mPatterns for Collaboration Between Data Scientists and Software Engineers
20mThe Need for Speed - AccelerateR-ing R Adoption in GSK