SkySense (Team Twisters)

Problem

People planning outdoor events, sports or travel get generic, unreliable forecasts rather than a personal answer for their place and time.

Solution

A web app that combines NASA Earth observation data with AI analysis for ultra-local predictions of rain, wind, temperature, humidity and visibility, real-time risk assessments and suggested safest time windows, plus a conversational assistant (TwisterBot) that sends adaptive alerts and reminders.

Demo moment

Not described in the sources.

Why it stood out

• organizer statedWon the award for the most innovative use of technology.Source ↗
• organizer statedTurns weather variables into a decision: a real-time risk assessment plus the safest time window for an activity.Source ↗
• team statedGrounded in named NASA sources (POWER API, Earthdata, Giovanni, GES DISC) alongside Open-Meteo, with a Gemini-powered assistant for alerts.Source ↗
• team statedSubmission package is complete: live product URL, demo video, data list and a specific AI-use disclosure (code assistance, design-to-code, icon generation) that separates the team's own work from tool help.Source ↗
• curator inferenceAnswers the challenge's playful question literally, for one person and one plan, which makes the product understandable in a sentence.

Lessons for your next hackathon

• Convert raw data into the one decision the user wants (when is it safe?).
• For everyday-data challenges, personal and local beats global and generic.
• Write the AI-use disclosure as carefully as the pitch: say what the tools did and what the team built.
• LATAM precedent: a team from Mexico won a global technology award.