Robber Ducky AI

Demo1
Demo2

Inspiration

"The best way to learn is to teach it yourself." We were inspired by the recent influx of interest in conversational AI models, notably due to ChatGPT. Rather than learning through asking questions, however, we wanted to introduce a platform where you learn through answering questions.

What it does

Our project helps people from all different backgrounds, from students to educators to professionals, to have a in-depth learning experience where we the goal is to not just know, but to understand. Through our chatbot model, users can learn by first inputting an explanation of the topic. The chatbot then responds with a question in order to prompt the reader into explaining more.

How we built it

Utilizing the resources and workshops provided by TreasureHacks, we were able to gain valuable insights on how to get started with building our project. We used a variety of technologies such as the JavaScript React library for the front-end development and OpenAI API for our conversational model.

Challenges we ran into

When working with our team members, we were all in different time zones, which means we all had different working hours and sometimes response times could be slow, making coordination difficult. We also had different skills, which initially posed a challenge since we had to figure out how to delegate tasks across our team. Due to time constraints, we are not able to provide a video.

Accomplishments that we're proud of

We overcame multiple hurdles, such as learning how to organize our workflow based on the different skills we all had in order to work together efficiently. We looked into topics we weren't familiar with and learned as much as we could in the given time period.

What we learned

We learned that when dealing with a conversational model, we need to use large amounts of training data to continuously evaluate and improve a chatbot's performance. We can also monitor user interactions to help identify any issues with the chatbot or response generation. Additionally, we have learned that it is important to design a chatbot with specific use cases in mind. With a clear vision, we can provide the chatbot with the appropriate knowledge and resources to perform its tasks optimally.

What's next for Robber Ducky AI

In the future, perhaps further development of the model could allow the chatbot to give feedback to the user's response. This way, the user will know how accurate and thorough their response is, creating a more personalized experience.