tAI | Devpost

Home page of tAI
Course Page
Logo and name
Assignments page
Grades page
Grade Breakdown page

Inspiration

This project was heavily inspired by the poor experience of software used in university to view our courses, specifically, Avenue (D2L). As university students, navigating through these platforms proved to be cumbersome and time-consuming, impacting our overall productivity and hindering our ability to learn effectively.

Faced with these challenges, we recognized the need for a streamlined and user-friendly solution to enhance the educational experience. Our goal was to develop a tool that not only addressed the difficulties we encountered but also provided a seamless and efficient way to access concise course information.

With a vision for an improved learning platform, our project aims to overcome the limitations of existing systems by focusing on user experience and quick accessibility to vital course details. Leveraging the growth of AI, our project focuses using AI as not a medium to do the work for students, but as an aid to further improve their learning experience online.

Our aspiration is to contribute to the enhancement of educational platforms, making them more intuitive, responsive, and tailored to the needs of students. By doing so, we believe we can positively impact the learning journey for students like ourselves, fostering a conducive environment for academic success and knowledge attainment.

What it does

Each course is equipped with its own personalized chatbot, creating a dynamic and responsive communication channel. This tailored approach ensures that students receive information and assistance that is directly relevant to their specific coursework. Whether it's generating practice questions for assessments or keeping students organized with updates on due dates and important announcements, the chatbot is a versatile companion for academic success.

In addition to academic support, tAI acts as a central hub for organizational updates. Students can effortlessly stay informed about due dates, assignment submissions, and crucial announcements. This ensures that important information is easily accessible, reducing the likelihood of missed deadlines and enhancing overall productivity.

The integration of tAI into the learning environment is aimed at enhancing students' overall learning experiences. By providing seamless interaction and unparalleled convenience, tAI becomes an indispensable tool for students looking to navigate their academic journey more efficiently. The platform's commitment to personalized communication, study assistance, and organizational support reflects our dedication to fostering an environment where students can thrive and achieve their academic goals.

How we built it

To bring our chatbot to life and enhance its capabilities, we harnessed the power of the Cohere API. Cohere played a pivotal role in empowering our chatbot to respond intelligently to user queries and effectively summarize course content material. Leveraging Cohere's advanced natural language processing capabilities, our chatbot not only understands the nuances of user inputs but also generates contextually relevant and coherent responses.

The user interface was crafted using HTMX, a cutting-edge library that extends HTML to facilitate dynamic and real-time updates, which formed the foundation of our interactive UI. This allowed us to create a responsive and engaging user interface that adapts to user interactions without the need for constant page reloads.

Furthermore, for our backend, FastAPI, a modern, fast, web framework for building APIs with Python 3.7+ based on standard Python type hints, served as the backend framework. Its asynchronous capabilities and efficient design enabled us to handle concurrent requests, ensuring a smooth and responsive chatbot experience.

Finally, Tailwind CSS, a utility-first CSS framework, was employed for styling the user interface. Its simplicity and utility-first approach allowed us to rapidly design and customize the UI, ensuring a visually appealing and user-friendly experience. The combination of Tailwind CSS and Jinja2, a modern and designer-friendly templating engine for Python, enabled us to dynamically render content on the server-side and present it in a cohesive manner.

Challenges we ran into

In terms of challenges, working with Cohere proved to be quite a challenge. While the API was very coherent and easy to read/follow, finding the right parameters to use was quite difficult. We had to test various different methods to get the prompts we wanted which also proved very challenging. Finally after many attempts we found the right parameters in order to get our project working as attended

Accomplishments that we're proud of

We take immense pride in the substantial progress achieved within the 24-hour timeframe of this project. Witnessing our initial vision transform into a tangible reality has been a source of great joy and satisfaction. The collaborative efforts of our team, fueled by dedication and creativity, have not only met but exceeded our expectations.

What we learned

We learnt a multitude of things, especially the stack we decided to use. For many of us, it was our first time using HTMX along with Flask to create a fully functional website. It was also most of our first time experiencing Cohere and using their API.

What's next for tAI

We truly believe AI will will only get better from now on. It is only its worse at this very moment so why not leverage its amazing capabilities and use it to further the learning of students for the future. We also understand how it can abused easily, however it is still a powerful tool we students should leverage while it is still young and fresh, to pave a path and create restrictions around them before it is too late. Some future features include: Reading and summarizing course content, etc.