Tabacout

Inspiration

One of team members personally is struggling with his smoking addiction. So we wanted to help find a solution that would help him quit.

What it does

It takes user age, state, average amount of cigarettes consumed in a day, years smoking, and an optional photo. This data is then parsed and displayed of 30 years with 10 year gaps, for an eye-opening realization.

How we built it

For the frontend, we used Next.js with tailwind css and ESlint for fixing problems. For the backend we used Next.js Routes, Gemini for processing images, and real-time CDC data that is getting updated to help with the statistics. We deployed the application on AWS Amplify.

Challenges we ran into

Making sure that the data was accurate for the statistics was a challenge because finding a sustainable, updated source was tough. Making sure the AI was fast enough for AWS Amplify was another issue we came across.

Accomplishments that we're proud of

We are proud of how the AI uses an "intensity" formula to figure out how dramatic the changes are to the users face. We are also proud of the fact we found real government data that could be incorporated into the project.

What we learned

We have learned how to deploy on AWS, along with some of the errors that come along with it, such as the timeouts and implementing AI keys. We also improved our skills on version control, we made branches, merged, and checkpointing.

What's next for Tabacout

For the future, we are planning on making our app more efficient when it comes to the time it takes for AI to create the image, more detailed data for users, a history for users to look back on past results and compare.

Built With

• amazon-web-services
• cdc
• eslint
• gemini
• github
• nextjs
• restapis
• tailwind
• tsx
• typescript