Team
Noel Dsouza (ndsouza) Rucha Kulkarni (rmkulkar) Krisha Bhambani (kbhamban) Harshit Mehrotra (hmehrotr)
Inspiration
We came across the wealth of data collected by satellites about the earth and wondered how that can be put to good use. A cause that all of us are passionate about is spreading awareness about climate change. Reconciling our passion with this intriguing data resource was our inspiration.
What it does
It teaches children about variation in environmental conditions like temperature, air quality, and vegetation across the globe using a gamified experience. The application prompts users to estimate these conditions in an area of their choice and then reveals the truth with some more background about it. The app also has a leaderboard for users to create usage streaks and foster a healthy learning environment.
How we built it
We first came up with the scope for our MVP which was to focus on educating about temperature and air quality. We chose Flutter as our development platform as it offers cross-platform support and great documentation. To fetch environmental information, we used freely available satellite data APIs. For the purpose of generating more information about the area of interest to the user, we used Open AI's GPT API.
Challenges we ran into
We initially struggled to come up with an idea that made a tangible impact using the wealth of satellite data available out there. For the implementation, we struggled to find an API that provided information about different environmental aspects in a map interface.
Accomplishments that we're proud of
We think that coming up with the idea was a moment of inspiration and actually implementing an MVP that has potential to impact kids' perspective towards the environment was an achievement that we are proud of.
What we learned
We learned how to narrow down scope of broad topics. We also learned how to reconcile technology with impact. It is possible to think too much about the technology which might deprecate the quality of the product if it is low impact. We learnt to avoid that.
What's next for Atlas Aware
First of all, we plan to fully integrate the APIs in automated manner for the temperature and air quality features. We also trained predictive models (code in notebook in the GitHub repo) to forecast temperature and vegetation changes over the next decade and plan to integrate this as a part of the knowledge we impart to the user at the end of a game flow.
Later, we intend to add support for vegetation and other environmental factors to Atlas Aware.
Built With
- dart
- flutter
- gpt
- machine-learning
- meteoblue
- python
Log in or sign up for Devpost to join the conversation.