Syntra

Join study groups to work with peers and AI!

Comment

Inspiration

We found that many students in universities weren't part of study group despite them being so beneficial. And even then, people found it hard to keep track of important dates and documents. We realized that if we built an application that can combine every aspect of creating and maintaining study groups, we could help students build meaningful connections and learn more, while also creating an environment on their device which can reliably connect them to their work and others.

What it does

You can join and invite people to study groups, schedule your meetings on a calendar, upload shared notes and resources to common groups, and use AI to summarize and generate questions, flashcards, summary or search about them! The home page combines everything into a neat layout that gives the most important information first. The AI tab allows you to generate flash cards, summaries, and questions about documents shared between specific study groups, along with searching through them. The study group tab allows you to review and customize your study groups, along with joining public ones! Documents allow you to upload your documents, download shared ones, and review their source. The calendar tab allows you to see when study group meetings are, and when document due dates are!

How we built it

We used Amazon's Bedrock and Lambda to incorporate a Claude-3-Haiku AI. We used Next, Tailwind, & Electron for developing our frontend and app, and AWS and Flask for our backend. We used a combination of Python, Typescript, and JavaScript and bit of Shell Scripting to bring our project to life

Challenges we ran into

Integrating everything together was our main challenge. Individually, we could all work together on one part of it pretty well, but combining it together required deep levels of communication and collaboration. Within our workflows, we had to be ready to hand off important sections to teammates, and have the data pipeline and file transfer synced together. We had a hard time transforming different types of uploaded documents into information we can use to generate questions from at each individual point (sharing between people, using a model on the data, etc), and at the point of putting it all together, which was only possible after a lot of time and teamwork.

Accomplishments that we're proud of

It works! That is amazing after 21 hours of work! We are also particularly proud that we were able to be able to input anything (.pdf, .docx, etc.) and still work with the AI, as well have the file transfer system to be so efficient.

What we learned

We learned how to work together as a team to efficiently deploy a project in such a short timeframe. We are also particularly proud of learning many new frameworks, APIs, and AI models that can be used for future projects, ranging from AWS to shell scripting to Flask.

What's next for Syntra.

We want to add more quality of life features such that any user is not experiencing bugs or being forced to follow a specific pattern. We then want to add more features for communication and collaborating between users, such as a chat or different ways to study asynchronously), to use AI in more unique ways to help our users learn, add more functionality to the calendar, and more!

Built With

Share this project:

Updates