EcoCommute

• 

  Thumbnail

• 

  Website favicon

Inspiration

As college students juggling academics and tight schedules, we have come to realize how often we prioritize convenience over sustainability when commuting. As climate change becomes increasingly urgent, we sought to create a practical solution that empowers individuals to reduce their carbon footprint without sacrificing time or simplicity. EcoCommute was born from the idea that eco-friendly choices should be effortless and even fun, using AI to guide better transportation habits.

What it does

EcoCommute is an AI-powered web app that helps users plan their daily commutes as smartly and sustainably as possible.

• Users select their starting point and destination via an interactive map.
  • EcoCommute recommends the optimal travel mode based on traffic, distance, time of day, and environmental impact.
  • It calculates CO₂ emissions and savings, suggests the cleanest route the user has taken in the past, and awards eco badges for good behavior.
  • Using Google’s Gemini API, the app generates a natural-language explanation of why that route and mode were chosen. Users can visualize their impact over time through simulations and easily accessible insights.

How we built it

React + Tailwind CSS for a clean, responsive frontend.

• FastAPI with SQLite for the backend, handling routing, emissions tracking, and badge logic.
• Google Maps API for real-time traffic data and geolocation.
• Gemini Flash 2.0 for natural language explanations.
• Leaflet for map interactivity.
• AI-based logic for recommendation, route scoring, and impact simulation.

Challenges we ran into

• Geolocation + AI integration: Linking real-world coordinates with a clean explanation using Gemini required careful API handling and prompt design.
• Sanitizing the inputs for the frontend and backend
• Tracking trip history: We needed to develop a lightweight yet scalable method for storing and retrieving commute logs to support personalized recommendations.

Accomplishments that we're proud of

• Integrating AI explanations to give users not just what route to take, but why.
• Designing a fun, badge-based experience that makes sustainability engaging.
• Building a full-stack product with map input, live traffic, and environmental simulation in less than 72 hours.
• Creating an architecture that’s ready for future expansion, like dashboards and mobile versions.

What we learned

• Creative ways to collaborate using GitHub and git
• How to have the frontend collect data from the backend and send data to the backend
• How to combine AI, APIs, and user design to solve a real-world problem.
• How to work with LLMs like Gemini in a practical application, especially around crafting helpful prompts.
• The importance of clear visuals and simplicity when dealing with climate data.

What's next for EcoCommute

Built With

• css3
• fastapi
• gemini
• google-maps
• python
• react
• sqlite
• tailwind
• typescript