Inspiration

We haven't found any truly open machine learning tutors that were available to everyone. They always had underlying costs or were complicated to use, and none had intuitive UX. We aimed to solve this through our beautiful UI and simple UX.

What it does

After signing in, you can create lesson plans for any topic of your choosing. It will generate 3-5 lessons about your topic, and present them to you in a intuitive way. We use a robust machine learning model that is able to create custom learning content for everyone.

How we built it

Because of the complexity of our application, we ended up all working on each part of the website at the same time. We finished the front end together, and then finished the backend together.

Challenges we ran into

We ran into challenges while implementing our login system. We learned how to implement phone number authentication using Twilio. We also ran into some errors prompt engineering, since it was the first time that we had used LLMs in our projects.

Accomplishments that we're proud of

We are super proud of our UI, and the functionality of our program, and are happy with how it turned out. But we believe that we have accomplished our goal of having the most intuitive UX possible.

What we learned

We learned a lot more about Supabase's API, as we used many features that we had never used before, such as Twilio's mobile authentication protocol. We also learned a lot about Gemini's API, as we used it for the first time this hackathon.

What's next for Astura

We plan to add knowledge quizzes, as they are proven to be super effective tools to increase your learning. We also plan to add a gallery of lessons that anyone can add to or use. The ease of use of our app can also be improved, through more ways to login, better models to generate lessons, and better UI design.

Built With

  • shadcn
  • svelte
  • sveltekit
  • tailwindcss
Share this project:

Updates