LearnAI

• 

  Learn AI logo

Inspiration

In an era of endless short-form content, it's easy to become overwhelmed. Our inspiration springs from a desire to use this attention-capturing format for good. We believe that short-form content is not just a trend but a powerful tool for education, one that can adapt to the fast-paced lifestyle of modern learners. Think TikTok, but for learning.

What it does

Imagine a personal learning assistant that knows exactly what piques your interest and challenges your intellect. That's what our platform offers. By leveraging ChatGPT, we create concise, informative articles on a wide array of topics, curated to the user's preferences and learning goals. But we don’t just stop at providing information; we engage the user further. Each article is accompanied by generated thought-provoking questions, designed to encourage critical thinking and deeper comprehension. These follow-up queries aren't just random; they're smartly generated to ensure that the learning objectives are met, turning passive reading into an active learning session.

How we built it

We intended on building it with the following tech stack: Node.js backend, SwiftUI frontend, and an ElephantSQL database. However, for our demo we transitioned to solely using SwiftUI, mock data, and communication with the ChatGPT API on the frontend for the presentation as we had issues with DNS propagation and SSL certificates when trying to communicate with a Node.js backend API server hosted on AWS.

Challenges we ran into

We ran into quite a few challenges while working on setting up an API server for our mobile app to interact with. The first of which was deploying the API server to AWS Elastic Beanstalk. We spent a lot of time debugging IAM roles, proper file structure, and configuring environment properties, but ended up deploying a working API server and PostgreSQL database. Unfortunately, our mobile application was not able to interact with the API server as XCode does not allow sending non-HTTPS requests, and our server did not have an SSL certificate. We tried purchasing a domain name and an SSL certificate, but it took too long for the certificate to be approved. However, we got around this roadblock by creating mock data and interacting directly with the ChatGPT API in the front-end to demonstrate all of the features of our app.

We also had to balance our time with planning versus building the app well. Specifically, we tried to move quickly from database and API design to implementation. This allowed us to catch this SSL certificate issue early and move towards a leaner, working MVP for the presentation.

Accomplishments that we're proud of

We're proud of the agility and problem-solving skills we demonstrated in navigating deployment challenges. Our ability to pivot and still present a working model with complex logic directly on the frontend within a limited timeframe is a testament to our team's resilience and adaptability.

What we learned

We learned the importance of early SSL certification in API server setup and gained valuable experience in integrating a mobile app with the ChatGPT API. We also deepened our understanding of SwiftUI for app development, learning to rapidly prototype and iterate on our design.

What's next for LearnAI

Our next steps involve securing the SSL certificate to enable HTTPS communication between our mobile app and the API server. This will allow us to move beyond mock data, implementing a robust system that tracks user interactions and preferences, fully harnessing the database's capabilities to personalize the educational experience. We will also work on a fully-fledged design for the mobile app that allows users to view past upvotes, share facts, and more.

Built With

• amazon-web-services
• elephantsql
• firebase
• javascript
• swift
• swiftui