Problem
- Seven people spent about 4 hours a day checking KMA forecasts and warnings by region and making notice images
- External weather APIs refreshed irregularly, so data looked different at every lookup
Company
Weather Decision System & BW Caster
An automation that cut bad-weather calls and notice images from 4 hours a day for 7 people to under a minute
Period | 2026.04 ~ 2026.05
Bad-weather calls and notices that took 7 people 4 hours a day
-
Daily manual work
Before7 people, about 4 hours a day
AfterOne click, under a minute
My Role
- Built a regional weather collection and decision system on public data and external APIs
- Linked KMA forecasts and warnings with Windy and ApiFlash to automate region selection, templates and images
- Handled irregular refresh cycles with Google Drive caching, 8 triggers, freshness checks and auto backfill/retry
Result
- Daily manual work reduced to one click, finished in under a minute
- Tuned API quota, nationwide thresholds and region selection on real data; integrated into the team AI Dashboard
Technical Decisions
Google Drive cache and freshness checks
External weather APIs refresh irregularly, so caching and freshness checks keep results consistent
Running in Production
- 8 triggers with automatic backfill and retry keep output flowing through API delays or failures
- API quota, nationwide thresholds and region selection tuned continuously on real data
