NewLink

Demo

StormHacks.2024.-.NewLink.Demo.mp4

What is it?

Imagine arriving at a new university in a new country where everything feels unfamiliar. The sights, the sounds, and even the way people interact are different. As a new student, you might wonder what to wear, how to ask questions, or how to make friends. Now imagine having a friend who understands exactly what you’re going through, shares your interests and is ready to help you navigate this new chapter.

Introducing... NewLink! 🎉

This app is the ultimate social compass for newcomers to SFU. It aims to be a one-stop place for newcomers to learn about the university, its values, and culture and connect with other people.

Features

• 🙋‍♂️ Ask questions to our SFU chatbot and FAQ
• 🤝 Connect with other people by showing interest in student-made events
• 🗓 View various courses offered by SFU

How to run ?

• Use npm install command on both the client and server folders
• Download the Llama 3 weights on server folder and store the model in a folder called /server/model. You can get the model from this URL.
• In the backend folder enter npm run devstart
• In the frontend folder enter npm start
• Now the project should be running on http://localhost:3000/

Tech Stack

• React.js for Frontend
• Express.js for Backend/API
• MongoDB for database
• LangChain for finetuning llm

Our primary focus was to build an LLM chatbot that is finetuned to the SFU information. For this, we wanted to use a local running instance of an LLM model like the new llama 3 rather than an open API key. We constructed our API from scratch!