Inspiration
Our inspiration came from our member Sudhit, who has been trying self-improvement apps to improve himself. Unfortunately, none of the apps he did were reinforcing or really practical. So this year at the hackathon, he strove to make an app to help those seeking to improve themselves.
What it does
This app creates a "Ditto" of yourself, where you have an AI clone that will start just like you and continuously improve from there on. In your case, you would have to do tasks that AI has made based on how close you are to your ideal self. The closer you are, the harder the tasks. These tasks can be completed by taking a picture of the objective, which gives you. Based on those tasks and your consistency, it will upgrade your stats. The AI clone is meant to be a guide on how you would improve if you kept sticking to your routine.
How we built it
We used Xcode to do the front end of the project, and we used VSCode for the backend. We also have a free Gemini 2.5 api key we used for the AI that makes the tasks and judges. We also have a database that runs off Render. We also had some 3d aspect into the app, which allowed us to make a loading screen.
Challenges we ran into
We ran into challenges when we didn't know how to add a camera and other permissions because it was our first time doing a project in Xcode that needed a camera. The task took a while and wasted a lot of our time, but we figured it out, and we were in a good position. Another challenge was one other team that came to us during the beginning of the competition, which kept bugging us. Even though they were our friends, it made us have a terrible start to everything.
Accomplishments that we're proud of
We are proud of building a working system where users can complete real-life tasks using the camera. The AI can review those tasks and track progress. We also like how the Ditto idea makes self-improvement feel more personal and engaging.
What we learned
We learned how to use Xcode for app development and how to handle camera permissions. We also learned how to connect an AI model to an app and use it in a meaningful way. As a team, we improved our communication and time management.
What's next for Ditto
Next, we want to improve the AI so that it gives more accurate and helpful feedback. We also want to add better customization, so users can set more specific goals. In the future, we plan to add reminders, streak tracking, and maybe a social feature so users can stay motivated together.
Log in or sign up for Devpost to join the conversation.