Inspiration
Making Resolutions is one of the most common things to do, however another just as common thing to do is to give up on them. Often times this comes down to a lack of proper planning. With a world that is every expanding it's use of AI from rudimentary tasks to extremely complex problems we were inspired to help resolve this issue.
What it does
The plan for the app was to create an app with a calendar layout that would allow users to use a input box to send a goal to the AI that would then create smaller tasks that would be more simple to complete making resolutions easier to achieve. It also includes a plus button that would allow user to enter their more mundane tasks alongside their goals.
How we built it
In our front end we created components such as buttons, navigation bar, tasks calendar and more. With these components we used Shadcn to format the front end to look according to our taste. We created a database in supabase in order to store our data for the calendar and created buttons to manually input mundane tasks, we included a input box in order for the user to input their goals to be processed by gemini, gemini would then use the functions created in order to add tasks into the calendar. We also added buttons to properly navigate between pages and different dates in the calendar.
Challenges we ran into
The first challenge we ran into was creating and implementing a database with supabase, the API key often was not working properly and required various fixes to allow it to be usable for all 3 of us creators. Another challenge we ran into was creating the calendar itself, certain functions didn't work properly and it was difficult at times to ensure that it display as we hoped it would. Another challenge we faced was implementing the AI, from getting API keys to creating methods to giving inputs to the AI there was constant issues that needed to be addressed in order for the AI to be able to be implemented. After the AI finally implemented properly, we ran into problems as we had to deploy our front end on cloudflare and our backend in Render, this caused some issues as we need to make sure that both were communicating properly. Finally, the AI itself required very specific prompts in order to function properly, this required the user to have be more specific in how they wanted their goals to be scheduled which took away a lot from the purpose of the App.
Accomplishments that we're proud of
Despite all the challenges it presented we were able to apply the AI to the app, for all problems we faced we were able to use our Research skills in order to understand possible issues and brainstorm solutions.
What we learned
We were able to expand our knowledge on the various tools we used and even became aware of other tools. We were able to grow in teamworking skills and learned to divide the work and study simultaneously allowing for work to be done effectively while also being able to learn more about effective ways to complete our goals.
What's next for TaskoAI
The several challenges we faced kept the final product away from our ideal final project however the knowledge we gain still grows and possible ways to improve are already being thought of. We want to improve the UI to ensure all options presented are available and work properly, add the ability to separate the data for different users so that the user experience will be appropriate for them.
Built With
- cloudflare
- express.js
- gemini
- git
- github
- javascript
- node.js
- postgresql
- react
- render
- shadcn
- supabase
- vite

Log in or sign up for Devpost to join the conversation.