Finding the best location for kitesurfing has been a struggle. The combination of wind, gear specs and rider weight are important in deciding the best beach for the day's surfing amongst alternatives. Checking manually weather conditions for various beaches to find the best one before a trip is cumbersome and laborious. We’ve decided to improve this process. Bali in Asia is a great destination for kite surfing with many beaches in different locations on the island.

What it does

Our app helps users to find either the best kite size for predicted weather conditions or picks the best beach locations for user with their own kites.

How we built it

Python, JavaScript, Google Maps, Cloudant, IBM Weather Insights, Analytics for Apache Spark, Bluemix.

Challenges we ran into

  • Bluemix Apache Spark setup.
  • Scheduling of jobs to pull requested data everyday

Accomplishments that we're proud of

Successfully connecting all components to make user experience as smooth as possible.

What we learned

Predictive machine learning models for Spark

What's next for Surf Finder

  • Deploy as a mobile app or as a chat bot
  • Augment the analytics back end
  • Improve the front end
  • Add more locations - kite surfing hotspots
  • Introduce social features that can help also fine tune the algorithms via user feedback and add non weather variables
Share this project: