Inspiration
Our inspiration for this app came from our common struggle to get academic guidance that was readily accessibly and personalized. We wanted to pursue an idea that would make it easy for a student to ask any question they would have regarding their academic standing, courses, scheduling and etc.
What it does
Our app UMazing takes insightful information from a UofM student's audit checklist, analyzes their academic trajectory, and uses this information to produce informative outputs by utilizing a database of all the EECS classes and Flex Techs at UofM at this time.
How we built it
This was a full-stack project and we separated this project into a front-end and back-end section. The front-end was responsible for creating multi-step tool calls to render streaming react components, and also utilized the OpenAI API to prompt engineer tailored responses to user queries in the context of course planning. The back end was responsible for processing the audit checklist, the catalog of all the EECS courses and Flex Techs, implementing APIs, creating 3 PostgreSQL databases, and the overall syncing of both front-end and back-end. Additionally, the project was hosted on Railway.
Challenges we ran into
Documentation on tooling was not optimal and precise, there was significant self-learning as we had to learn on the fly on how to read and utilize the documentation that we read. We also had to learn how to use the OpenAI API to create context for our GPT in creating responses that we bellieved would be optimal for our app. The AI SDK that we used separated react server component state for AI components for client side. The document was lacking and this forced us to very precisely read all the documentation.
Accomplishments that we're proud of
We were very proud of getting a fully functional full-stack project in 24 hours for our first hackathon. Additionally, this idea was something that we all were passionate about and bringing it to life within such a short time span was incredibly exciting.
What we learned
When push comes to shove, I think every group member was able to overcome a significant challenge and produce an app from ideation to fruition.
What's next for UMazing
UMazing is going to get a lot of valuable feedback from other hackers and mentors and will be improved on in the future.
Built With
- next.js
- node.js
- openai
- postgresql
- python
- railway
- react
- typescript
- vercel
Log in or sign up for Devpost to join the conversation.