LearnSpace

Inspiration

We had the the thought when we thought to ourselves "what keeps me from studying?" And that's when it hit us. The continuous reels and feeds we scroll through at our leisure, a.k.a, "Doom-Scrolling" So we figured if we could turn Doom-Scrolling into a way of learning, we could both combat that habit as well as provide a learning resource.

What it does

LearnSpace is a web application that allows users to submit PDFs they wish to learn from or study. It has the design and aesthetic of a social media app, but instead of posts there are questions on those subjects of interests, and instead of a field to reply/comment, a field to answer the provided question.

How we built it

We first came together and structured our project into 3 main sections. The frontend, the backend, and the NeuralSeek API service. With those clear separations we were able to divide the tasks and work on them at the same time to maximize our speed. The frontend is built using NEXT.js, with Axios to handle communication to the backend. The backend was built in Python, using the FastAPI library. To store persistent data in the backend we used a Supabase (Postgres) database, and Auth0 to handle user sessions and authorization. Lastly, the NeuralSeek API service we used allowed us to orchestrate complex agent operations, allowing us to generate unique questions from LLMs and automate grading of short response questions.

Challenges we ran into

One of our greatest challenges was losing a team member early into the competition. He had become unwell, and had to withdraw from working. However, we pulled through. Each of us sharing the burden and working late into the night.

Another challenge we ran into was wrangling these unfamiliar services within the short time provided for the Hackathon. We struggled hard but were able to come to a decent understanding of these services and how to use them in conjunction.

Accomplishments that we're proud of

We managed to complete an MVP within the time constraints of the Hackathon while down a team member. And not only that, but while using unfamiliar services for the first time. Our fatigue and elation at submitting on time is a testament to our dedication.

What we learned

We learned what it means to work as a team that relies on each other. With time as short as it was, there wasn't much time to check over someone else's code and "grade" them. Instead, we each focus on our own parts, helping each other when they need it, and trusting the results of our peers as equals. Of course there are times when something is wrong, but respectfully pointing it out for each other and accepting such critique only makes us better programmers.

What's next for LearnSpace

LearnSpace is an infant compared to full-fledged products. There is so much room for improvement and growth that could really elevate LearnSpace to an amazing service. Only time will tell where LearnSpace goes from here.