Loading page...Загружаем страницу...正在加载页面...
AtlasRepo

Nice Out: an app that learns your kind of nice weather, then tells you to go outside

Description and use

Nice Out learns individual weather preferences from a handful of ratings and recommends optimal outdoor windows, refining predictions as users log outings.

Problem

Weather apps give everyone the same numbers, so users cannot decide when to go outside based on personal comfort.

Outcome

The app delivers a single, personalized recommendation for the next outdoor window, with a 95% chance of enjoyment, and improves over time as users log outings.

Workflow

  • Pick your area (location rounded to ~5 km)
  • Rate 14 real hours from your past week (yes/no to "would you have gone out?" )
  • TabPFN learns your preference line and scores every hour of the next 36 hours
  • App shows the next optimal window with likelihood and key conditions
  • Go outside and tap Great/Fine/Bad to log the outing
  • Each outing becomes a new training row for TabPFN
  • Ask the Mastra agent about a specific plan (e.g., "When can I do a 1-hour run tomorrow?")

Evidence

Nice Out learns what "nice out" means to you: Pick your area. No account. Your location is rounded to about 5 km.

Source

Rate 14 real hours from your past week. "Would you have gone out?" Yes or no.

Source

TabPFN learns your line from those few rows and scores every hour of the next 36.

Source

The app also stores the chance it showed for that hour before you went. After 5 outings, it shows how many it called right.

Source

  • Record date: 2026-10-06T05:41:06.427Z