Keynote Speaker-Isabel Zimmerman
summary
Compares pandas to polars (Rust backend) and claims a simple switch speeds common workflows, describes building an MLOps package (Vetiver‑style) and an IDE called Positron IDE with variables/plots panes, a permanent console, remote SSH, and a data explorer that handles millions of rows and can convert to code so GUI edits become reproducible scripts; demos that run buttons invoke streamlit run, render‑on‑save avoids restarts, and a help pane surfaces docstrings — features are justified by user testing and live feedback.
AI-written summary · about
chapters (14)
- 0:00Introduction
- 1:49Why tools matter in data science
- 3:53Everyone is a tool builder
- 8:28Hard skills of great tools
- 10:17Positron IDE
- 11:44Reproducibility in tool design
- 14:30Balancing flexibility in tools
- 17:40AI and automation limits
- 19:24Soft skills for user-centered tools
- 22:31Discoverability and hidden features
- 26:04Small improvements with big impact
- 28:32Explaining your work
- 33:50Building tools holistically
- 36:18Q&A
