Inspiration
We were ambitious to learn more about how generative AI can be used as a tool with assessing risk and remediation. We originally were planning to make a project around 911 operation but then decided to go with AI image analysis because its cooler in our opinion, and its more user friendly.
What it does
Our app takes user uploaded images about a room they are interested for preparing for natural disasters. Then we use selective prompting and generative AI on the received image. Then we give back the user a risk analysis report about risks in the room. We also give them with a check list and a list of improvements to make in the room.
How we built it
So we used mainly TypeScript and React for the front-end and Python and FastAPI for the back-end. We used our client to send over image data and possible location data and we used our server to call Google's Gemini AI model for processing and parsing. Then the client receives the risk analysis from the server with the checklist.
Challenges we ran into
Trying to stay to finish our goals was not easy. Working with new technologies to us like Docker and back-end frameworks like FastAPI. Also prompting was a big one.
Accomplishments that we're proud of
We achieved our goals despite not knowing much. Communicating across multiple APIs was also very exciting.
What we learned
How to properly plan and design a project, and using new tools like FastAPI.
What's next for RoomReady
We plan on hosting the project for interested people to view. Also add more AI models to generate better analyses.
Log in or sign up for Devpost to join the conversation.