Build Simple and Scalable Apps with Shiny
summary
Claims that Shiny for Python infers a reactive computation graph and builds it by runtime tracing so only affected outputs re-render. Shows concrete primitives: reactive calculations with automatic caching for shared work, reactive events/effects for side effects, lazy evaluation for hidden tabs, and invalidation semantics when upstream inputs change. Demonstrations include a sampling-and-filtering pipeline that reuses one database sample and a UI that swaps dependencies so only minimal recomputation runs — practical examples of scaling prototype to product.
AI-written summary · about
chapters (10)
- 0:00Shiny for data scientists
- 3:11Shiny solves last mile problem
- 7:47How Shiny infers reactivity
- 12:43Transparent reactivity in Shiny
- 15:31Scaling Shiny with reactivity
- 22:30Efficiency with reactive calculations
- 24:15Event programming in Shiny
- 27:45Reactive effects vs calculations
- 28:48Shiny for growing apps
- 29:30Q&A
related talks
25mUnderstanding reactive execution in Shiny
44mShiny: Data-centric web applications in Python
44mShiny for Python: Interactive apps and dashboards made easy-ish
27mShiny: A GUI for the Tidyverse
31mTips to Level-Up Your Shiny for Python Applications
20mRunning Shiny without a server
50mShiny in production: Principles, practices, and tools
20mWeb applications with Shiny and React and AI